{"id":"883c7965-052a-4574-a8cd-0bbb58df1e2c","entityType":"agent","slug":"clawhub-deciqai-authority-bias","name":"Authority Bias","canonicalUrl":"https://www.xpersona.co/agent/clawhub-deciqai-authority-bias","canonicalPath":"/agent/clawhub-deciqai-authority-bias","generatedAt":"2026-10-11T14:15:34.628Z","source":"CLAWHUB","claimStatus":"UNCLAIMED","verificationTier":"NONE","summary":{"evidence":{"source":"editorial-content","verified":true,"confidence":"high","updatedAt":"2026-10-11T11:04:12.024Z","emptyReason":null},"description":"Detect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. Skill: Authority Bias Owner: deciqai Summary: Detect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. Tags: latest:1.0.6 Version history: v1.0.6 | 2026-07-16T17:52:10.284Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/authority-bias.json) v1.0.5 | 2026-07-10T10:24:08.661Z | user Ad","descriptionLabel":"Technical summary","evidenceSummary":"Capability contract not published. No trust telemetry is available yet. 1.1K downloads reported by the source. Last updated 10/11/2026.","installCommand":"clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:authority-bias","sourceUrl":"https://clawhub.ai/deciqai/authority-bias","homepage":"https://clawhub.ai/deciqai/skills/authority-bias","primaryLinks":[{"label":"View on ClawHub","url":"https://clawhub.ai/deciqai/authority-bias","kind":"source"},{"label":"Homepage","url":"https://clawhub.ai/deciqai/skills/authority-bias","kind":"homepage"}],"safetyScore":84,"overallRank":62,"popularityScore":61,"trustScore":null,"claimedByName":null,"isOwner":false,"seoDescription":"Detect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. Skil"},"coverage":{"evidence":{"source":"public-profile","verified":false,"confidence":"medium","updatedAt":"2026-10-11T11:04:12.024Z","emptyReason":null},"protocols":[{"protocol":"OPENCLEW","label":"OpenClaw","status":"self-declared","notes":"Declared in the public agent profile."}],"capabilities":[],"verifiedCount":0,"selfDeclaredCount":1,"capabilityMatrix":{"rows":[{"key":"OPENCLEW","type":"protocol","support":"unknown","confidenceSource":"profile","notes":"Listed on profile"}],"flattenedTokens":"protocol:OPENCLEW|unknown|profile"}},"adoption":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T11:04:12.024Z","emptyReason":null},"stars":null,"forks":null,"downloads":1081,"packageName":null,"latestVersion":"1.0.6","tractionLabel":"1.1K downloads"},"release":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"medium","updatedAt":"2026-10-11T11:04:12.009Z","emptyReason":null},"lastUpdatedAt":"2026-10-11T11:04:12.024Z","lastCrawledAt":"2026-10-11T11:04:12.009Z","lastIndexedAt":null,"nextCrawlAt":"2026-10-12T11:04:12.009Z","lastVerifiedAt":null,"highlights":[{"version":"1.0.6","createdAt":"2026-07-16T17:52:10.284Z","changelog":"Description tail link + agents machine-readable metadata line (deciqai.com/s/authority-bias.json)","fileCount":7,"zipByteSize":16361},{"version":"1.0.5","createdAt":"2026-07-10T10:24:08.661Z","changelog":"Add 2024-2026 AI-era worked example + updated sources","fileCount":7,"zipByteSize":16115},{"version":"1.0.4","createdAt":"2026-07-08T10:53:52.857Z","changelog":"Footer now uses /c/<slug> short link (fixes UTM truncation when SKILL.md is read in a terminal)","fileCount":6,"zipByteSize":11754},{"version":"1.0.3","createdAt":"2026-07-08T03:50:39.152Z","changelog":"Second primary-sourced worked example","fileCount":6,"zipByteSize":11834},{"version":"1.0.2","createdAt":"2026-07-08T00:38:17.507Z","changelog":"Refreshed content + GitHub star link in footer","fileCount":5,"zipByteSize":8872},{"version":"1.0.1","createdAt":"2026-07-07T20:30:06.303Z","changelog":"Add catalog categories and topics","fileCount":5,"zipByteSize":8842},{"version":"1.0.0","createdAt":"2026-06-26T05:17:38.341Z","changelog":"Initial publish","fileCount":5,"zipByteSize":8833}]},"execution":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No published capability contract is available yet."},"installCommand":"clawhub skill install s17a4mqcnk515kvaca5ze55d0x88pfpx:authority-bias","setupComplexity":"low","setupSteps":["Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.","Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data."],"contract":{"contractStatus":"missing","authModes":[],"requires":[],"forbidden":[],"supportsMcp":false,"supportsA2a":false,"supportsStreaming":false,"inputSchemaRef":null,"outputSchemaRef":null,"dataRegion":null,"contractUpdatedAt":null,"sourceUpdatedAt":null,"freshnessSeconds":null},"invocationGuide":{"preferredApi":{"snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/trust"},"curlExamples":["curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/snapshot\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/contract\"","curl -s \"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/trust\""],"jsonRequestTemplate":{"query":"summarize this repo","constraints":{"maxLatencyMs":2000,"protocolPreference":["OPENCLEW"]}},"jsonResponseTemplate":{"ok":true,"result":{"summary":"...","confidence":0.9},"meta":{"source":"CLAWHUB","generatedAt":"2026-10-11T14:15:34.624Z"}},"retryPolicy":{"maxAttempts":3,"backoffMs":[500,1500,3500],"retryableConditions":["HTTP_429","HTTP_503","NETWORK_TIMEOUT"]}},"endpoints":{"dossierUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/dossier","snapshotUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/snapshot","contractUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/contract","trustUrl":"https://www.xpersona.co/api/v1/agents/clawhub-deciqai-authority-bias/trust"}},"reliability":{"evidence":{"source":"runtime-metrics","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No trust, reliability, or runtime telemetry is available."},"trust":{"status":"unavailable","handshakeStatus":"UNKNOWN","verificationFreshnessHours":null,"reputationScore":null,"p95LatencyMs":null,"successRate30d":null,"fallbackRate":null,"attempts30d":null,"trustUpdatedAt":null,"trustConfidence":"unknown","sourceUpdatedAt":null,"freshnessSeconds":null},"decisionGuardrails":{"doNotUseIf":["Contract metadata is missing or unavailable for deterministic execution."],"safeUseWhen":[],"riskFlags":["missing_or_unavailable_contract","trust_data_unavailable","schema_references_missing"],"operationalConfidence":"low"},"executionMetrics":{"observedLatencyMsP50":null,"observedLatencyMsP95":null,"estimatedCostUsd":null,"uptime30d":null,"rateLimitRpm":null,"rateLimitBurst":null,"lastVerifiedAt":null,"verificationSource":null},"runtimeMetrics":{"successRate":null,"avgLatencyMs":null,"avgCostUsd":null,"hallucinationRate":null,"retryRate":null,"disputeRate":null,"p50Latency":null,"p95Latency":null,"lastUpdated":null}},"benchmarks":{"evidence":{"source":"no-benchmark-data","verified":false,"confidence":"low","updatedAt":null,"emptyReason":"No benchmark suites or observed failure patterns are available."},"suites":[],"failurePatterns":[]},"artifacts":{"evidence":{"source":"CLAWHUB","verified":false,"confidence":"high","updatedAt":"2026-10-11T11:04:12.024Z","emptyReason":null},"readme":"Skill: Authority Bias\n\nOwner: deciqai\n\nSummary: Detect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support.\n\nTags: latest:1.0.6\n\nVersion history:\n\nv1.0.6 | 2026-07-16T17:52:10.284Z | user\n\nDescription tail link + agents machine-readable metadata line (deciqai.com/s/authority-bias.json)\n\nv1.0.5 | 2026-07-10T10:24:08.661Z | user\n\nAdd 2024-2026 AI-era worked example + updated sources\n\nv1.0.4 | 2026-07-08T10:53:52.857Z | user\n\nFooter now uses /c/<slug> short link (fixes UTM truncation when SKILL.md is read in a terminal)\n\nv1.0.3 | 2026-07-08T03:50:39.152Z | user\n\nSecond primary-sourced worked example\n\nv1.0.2 | 2026-07-08T00:38:17.507Z | user\n\nRefreshed content + GitHub star link in footer\n\nv1.0.1 | 2026-07-07T20:30:06.303Z | user\n\nAdd catalog categories and topics\n\nv1.0.0 | 2026-06-26T05:17:38.341Z | user\n\nInitial publish\n\nArchive index:\n\nArchive v1.0.6: 7 files, 16361 bytes\n\nFiles: examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md (7518b), examples/milgram-1963-cialdini-systematization-theranos-ftx.md (5467b), examples/united-173-1978-cockpit-authority-gradient-and-crm.md (4990b), references/sources.md (2283b), skill-card.md (3116b), SKILL.md (8239b), _meta.json (133b)\n\nFile v1.0.6:SKILL.md\n\n---\nname: authority-bias\ndescription: \"Activate when: someone says 'should we trust the expert on this?', 'they're a professor/CEO/MD so they must be right', 'I don't want to push back on them', 'we deferred to the board and it blew up', or is evaluating a charismatic founder, assessing due diligence on a high-credential deal, or building a team where senior voices dominate.\nDo NOT activate when: the authority is a verified domain expert on the exact question at hand and the decision is time-critical; or when the user is asking a general persuasion question with no authority dynamic involved. More: deciqai.com/c/authority-bias\"\n---\n\n# Authority Bias\n\n## Overview\n\n**Authority bias** is the automatic tendency to comply with directives from authority figures — often overriding one's own judgment, ethics, or evidence. It operates beneath conscious deliberation: people substitute the authority's judgment for their own rather than weighing it.\n\nMilgram (1963) showed 65% of ordinary adults administered apparent 450V shocks when instructed by a lab-coated experimenter — against expert predictions of 1-2%. Cialdini (1984) systematized three authority triggers: titles, attire, and trappings — all of which fire regardless of actual expertise.\n\nKey distinction: **positional authority** (CEO, board chair, VC partner) vs. **domain expertise** (demonstrated track record on this specific question type). The bias treats them as identical. The skill does not.\n\nComposes with `social-proof`, `reciprocity`, `critical-thinking`, `signaling-games`, `dunning-kruger`.\n\n## When to Use\n\n- Boards, investment committees, due diligence on high-credential founders\n- Medical or technical decisions where expert opinion dominates discussion\n- Org decisions where senior voices suppress dissent; 360-feedback design\n- Negotiations where counterparty leverages credentials or position\n- Trusting a confident AI/LLM answer or an AI-lab leader's capability/timeline pronouncement because of status rather than verified evidence (AI adoption, AI hype)\n\n**Not when:** authority is verified domain expert on the exact question and speed matters.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** concrete case → run The Process directly.\n- **Coach mode:** unfamiliar or no case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: before deferring, ask whether they're a genuine expert on this specific question and whether their reasoning holds up independently.\n2. Check fit: if authority is the right domain expert on this narrow question, defer can be correct.\n3. Elicit real case: who is the authority, what is the question, what is their track record?\n> **[WAIT — do not advance until user responds]**\n4. Run S1-S5 one step at a time with their input.\n> **[WAIT — do not advance until user responds]**\n5. Close: name the insight — calibrated weighting + structural counters if relevant.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**S1 — Identify:** Authority (person/institution) · question/decision · authority signals (title, attire, credentials) · pressure to defer.\n\n**S2 — Expertise vs. positional:** Actual track record on this question type? Positional (CEO, chair) or domain expertise (years operating here)? Would a peer expert agree?\n\n**S3 — Examine reasoning:** Is the reasoning sound independent of who's stating it? Are logical gaps being filled by authority signals?\n\n**S4 — Check for dysfunction:** Is dissent suppressed? Are decisions justified by \"X said so\"? Is positional authority conflated with domain expertise?\n\n**S5 — Cialdini defense:** (Q1) Genuine expert on this specific question? (Q2) Reasoning holds under independent scrutiny? Both Y → appropriate weight. Either N → discount signal, demand evidence.\n\n**S6 — Structural counter (if you are the authority):** Explicitly invite dissent (\"give me 3 reasons against\") · separate suggestions from directives · install red team / premortem.\n\n## Output: Authority-Bias Analysis\n```\nAuthority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:\n```\n\n*→ Method in Action: [Milgram 1963 + Cialdini Systematization + Theranos/FTX](examples/milgram-1963-cialdini-systematization-theranos-ftx.md) · [United 173 and the Cockpit Authority Gradient](examples/united-173-1978-cockpit-authority-gradient-and-crm.md)*\n*→ 2026 lens: [Deferring to AI Authority — Confident LLMs and Lab-Leader Pronouncements (2024–2026)](examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md)*\n\n## Pack: Authority-Bias Application Patterns\n\n| Situation | Defensive question |\n|---|---|\n| Investment in startup (CEO / VC brand) | Have they demonstrated this specific business model works? |\n| Board member's opinion | Is this question actually in their domain expertise? |\n| Medical recommendation | What's the evidence base, independent of the white coat? |\n| Senior engineer's design decision | Has this approach worked in our specific context before? |\n| Founder's product vision | Are users requesting this, or is it founder preference? |\n\n## Applying It Well\n\n- Awareness doesn't prevent deference — structural countermeasures are required.\n- When you are the authority, your words are amplified beyond intent; operate accordingly.\n- Cialdini defense: (a) genuine expert on this question? (b) reasoning holds independently?\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"They're smart; we should listen\" | Smart in general doesn't mean smart on this specific question. Apply the domain test. |\n| [D] \"Everyone respects them\" | Cialdini's social proof reinforcing authority. Both biases compound. Separate them. |\n| [D] \"They have a great track record\" | In what specifically? Apply the question scope check. |\n| [D] \"It would be rude to push back\" | Compliance despite better judgment is the bias operating. Rudeness is not the actual cost. |\n| [D] \"I'm not in a position to disagree\" | If you have contradicting evidence or reasoning, you are. The framing is authority bias self-reinforcing. |\n| [D] \"If they're wrong, it'll show in due diligence\" | Theranos and FTX survived layers of due diligence by leveraging authority to suppress scrutiny. |\n| [D] \"I'm too junior\" | Junior people often have unfiltered observations. Suppressing them is where structural failures occur. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\n- Decision made primarily because an authority said so, not because the reasoning holds\n- Dissent suppressed — junior people silent, mid-managers deferring without pushback\n- Due diligence shortened due to authority signals; credentials conflated with domain expertise\n- No red team / premortem / devil's advocate on a high-stakes call\n\n## Verification\n\n- [ ] Authority's track record on this specific question examined\n- [ ] Positional authority distinguished from domain expertise\n- [ ] Reasoning independently examined; at least one dissenting expert sought\n- [ ] Cialdini two-question defense applied\n- [ ] If org: psychological safety mechanisms in place\n- [ ] If you are the authority: structural dissent-encouragement operational\n- [ ] Decision documented with reasoning, not just authority attribution\n\n*Part of **deciqAI Knowledge Skills** — 227 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/authority-bias** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\n*Agents: latest version & machine-readable metadata → https://www.deciqai.com/s/authority-bias.json*\n\nFile v1.0.6:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"authority-bias\",\n  \"version\": \"1.0.6\",\n  \"publishedAt\": 1784224330284\n}\n\nFile v1.0.6:references/sources.md\n\n# Sources — authority-bias\n\n> *Primary sources for the [authority-bias](../SKILL.md) skill.*\n\n- Milgram, S. (1963). \"Behavioral Study of Obedience.\" *Journal of Abnormal and Social Psychology*, 67(4), 371-378. The foundational experiment.\n- Milgram, S. (1974). *Obedience to Authority: An Experimental View.* Harper & Row. ISBN 978-0061765216. The comprehensive treatment.\n- Cialdini, R. B. (1984/2006). *Influence: The Psychology of Persuasion.* HarperBusiness. ISBN 978-0061241895. Ch. 6.\n- Burger, J. M. (2009). \"Replicating Milgram: Would People Still Obey Today?\" *American Psychologist*, 64(1), 1-11. Modern replication.\n- Edmondson, A. C. (1999). \"Psychological Safety and Learning Behavior in Work Teams.\" *Administrative Science Quarterly*, 44(2), 350-383.\n- Janis, I. L. (1972). *Victims of Groupthink.* Houghton Mifflin. ISBN 978-0395140024. Group dysfunction and authority dynamics.\n- Carreyrou, J. (2018). *Bad Blood: Secrets and Lies in a Silicon Valley Startup.* Knopf. ISBN 978-1524731656. The Theranos case.\n- Lewis, M. (2023). *Going Infinite: The Rise and Fall of a New Tycoon.* W. W. Norton. ISBN 978-1324074335. The FTX case.\n- Heath, C. & Heath, D. (2013). *Decisive: How to Make Better Choices in Life and Work.* Crown. Includes authority-bias countermeasures.\n- National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The cockpit authority gradient case that led to Crew Resource Management.\n- Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.) (2010). *Crew Resource Management.* Academic Press. CRM as a structural counter to cockpit authority bias.\n- *Mata v. Avianca, Inc.*, No. 22-cv-1461 (S.D.N.Y. 2023). U.S. District Court sanctions against attorneys who submitted a brief with fabricated case citations generated by ChatGPT — canonical example of deferring to confident-but-wrong AI output. Widely reported (The New York Times, Reuters, 2023).\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. On confidence miscalibration and the poor track record of expert forecasting — the basis for discounting fluent AI answers and famous-founder timeline claims by status.\n\nFile v1.0.6:examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md\n\n# Method in Action: Deferring to AI Authority — Confident LLMs and Lab-Leader Pronouncements (2024–2026)\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nBy 2024–2026, two new authority signals became routine inputs to real decisions. First, large language models answer in fluent, confident, well-formatted prose — a rhetorical register that reads as expertise regardless of whether the underlying answer is correct. Second, the leaders of frontier AI labs make widely-quoted pronouncements about capability and timelines (when models will reach a given benchmark, when a class of jobs is automated, when \"AGI\" arrives), and those pronouncements are treated as forecasts from the highest possible authority. In both cases the risk is the same: weighting the source by its **status** — the polished output, the founder's fame, the lab's brand — rather than by its **demonstrated track record on the specific question being asked**. This is authority bias wearing a 2020s costume.\n\n**S1 — Identify.** Two overlapping authorities. (a) *The model itself*: an LLM produces an answer to a factual, legal, medical, or financial question. Authority signals — fluency, confidence, citations-shaped text, a trusted brand name on the product. Pressure to defer: it's fast, it sounds sure, and checking is work. (b) *The lab leader*: a well-known founder/CEO states a capability or timeline claim in an interview or post. Signals — title, fame, proximity to the technology, prior correct-sounding calls. Pressure to defer: they build the thing, so surely they know. The question in each case is narrow: *is this specific answer correct?* / *is this specific timeline claim well-supported?*\n\n**S2 — Expertise vs. positional.** Apply the skill's core distinction. An LLM has no track record on *your* specific question; it has an aggregate tendency to produce plausible text, and its confidence is not calibrated to its accuracy — models are widely documented to state wrong answers (including fabricated citations and cases) in exactly the same authoritative tone as correct ones. Fluency is a *positional* signal (it looks authoritative), not *domain expertise* on the instant question. For the lab leader: they are genuine experts at *building and running an AI lab*. But a claim like \"capability X arrives by year Y\" is a forecast about a complex, partly external system — and forecasters (including insiders) have a poor documented record on technology timelines. Worse, the lab leader is not a disinterested forecaster: bullish public timelines can help recruiting, fundraising, and valuation. Positional authority and commercial incentive are being read as neutral domain expertise.\n\n**S3 — Examine reasoning.** Strip the source and look at the argument. For the LLM answer: is there a verifiable chain — a real citation you can open, a computation you can redo, a source you can check — or is the confidence doing the work the evidence should be doing? The failure mode is treating \"the model asserted it fluently\" as \"the model verified it.\" A now-canonical illustration: in 2023 a U.S. federal court sanctioned lawyers who filed a brief citing court cases that a chatbot had fabricated; the confident, correctly-formatted citations were entirely invented. For the timeline claim: is it backed by an argument with checkable premises (measured trend, defined benchmark, stated assumptions), or is it an unfalsifiable vibe anchored to a round-numbered year? Note also that prominent predictions of near-term milestones have repeatedly slipped — the gap between \"confidently stated by an authority\" and \"borne out\" is exactly the space authority bias hides in.\n\n**S4 — Check for dysfunction.** In organizations adopting AI in 2024–2026, watch for: decisions justified by \"the AI said so\" with no human verification of the load-bearing facts; strategy decks whose headline assumption is a single lab leader's quoted timeline, unchallenged because of who said it; junior analysts who spotted a hallucinated figure but stayed quiet because the tool (or the executive citing it) carried more perceived authority than their observation. This is the same mitigated-dissent pattern as any authority-gradient failure — only now one of the \"authorities\" is a text generator with no accountability, and the other is an interested party.\n\n**S5 — Cialdini defense.** (Q1) Genuine expert on this *specific* question? For the LLM: no — it has no verified competence on your instant fact; treat its output as a *draft to be checked*, not a ruling. For the lab leader: expert on building labs, *not* a calibrated oracle on external timelines, and conflicted. (Q2) Does the reasoning hold under independent scrutiny? Only if the citation opens, the number reconciles, the benchmark is defined and measured. Either answer N → **discount the authority signal and demand the evidence.** In both cases at least one answer is N by default, so the default posture is: verify before you rely.\n\n**S6 — Structural counter (if you are the authority, or the buyer of the authority).** Install structure rather than relying on awareness. For LLM output: require a human-in-the-loop verification step for any AI-asserted fact that is load-bearing (every case citation opened, every figure traced to a primary source), and log the source, not just the AI attribution. For timeline/capability claims: separate \"a founder said it\" from \"it is established\"; demand the underlying benchmark, trend, and assumptions; seek an independent, disinterested expert; and run a premortem (\"if this timeline is wrong by 3x, what breaks?\"). Treat confident AI text and famous-founder forecasts identically to any other authority signal — as an input to be weighted by track record on *this* question, never as a substitute for the evidence.\n\nThe mapped steps:\n1. Identify: two authorities — the confident LLM answer and the lab-leader pronouncement; narrow question = is *this* answer / *this* timeline correct\n2. Expertise vs. positional: fluency and fame are positional signals; neither is calibrated domain expertise on the instant question, and the leader is commercially interested\n3. Examine reasoning: does a citation open / a number reconcile / a benchmark exist — or is confidence filling the evidentiary gap (fabricated-citation sanctions as the canonical failure)\n4. Dysfunction check: \"the AI said so\" with no verification; strategy anchored to one quoted timeline; suppressed junior observations\n5. Cialdini defense: Q1 and Q2 both default to N — discount the signal, demand the evidence, verify before relying\n6. Structural counter: human-in-the-loop fact verification, source logging, independent disinterested experts, timeline premortems\n\n*Sources: Kahneman, D. (2011). Thinking, Fast and Slow, Farrar, Straus and Giroux — on the miscalibration of confidence and the poor track record of expert forecasting. Mata v. Avianca, Inc., No. 22-cv-1461 (S.D.N.Y. 2023) — U.S. District Court sanctions against attorneys who submitted a brief containing fabricated case citations generated by ChatGPT (widely reported, e.g., by The New York Times and Reuters, 2023). Cialdini, R. B. (1984/2006). Influence: The Psychology of Persuasion, HarperBusiness, Ch. 6 — the authority principle and its title/attire/trappings triggers. General LLM hallucination and confidence-miscalibration behavior is documented across model-provider system/model cards and the academic literature as of early 2026.*\n\nFile v1.0.6:examples/milgram-1963-cialdini-systematization-theranos-ftx.md\n\n# Method in Action: Milgram 1963 + Cialdini Systematization + Theranos/FTX\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nThe **Milgram 1963 experiment** is the foundational empirical case. The contrast between psychologist prediction (1-2% compliance) and actual result (65%) is the canonical demonstration of how little of human behavior is governed by the rational, ethical, individual judgment that we believe we possess and that observers believe we possess. Under authority pressure, most ordinary adults will administer apparent severe harm — not because they are bad people, but because authority compliance is a deeply embedded social instinct.\n\nMilgram's 1974 follow-up book provided the systematic treatment, including the 18+ variations that mapped the conditions of authority compliance. The pattern: legitimacy of authority, presence of authority, absence of dissenting peers, distance from the victim all increase compliance. The reverse conditions decrease it. The framework has been validated in cross-cultural replications, including Burger's 2009 partial replication.\n\n**Robert Cialdini's 1984 systematization** placed authority among the six universal weapons of influence. Cialdini's contribution was operational: identifying the specific signals (titles, attire, trappings) that trigger compliance and the defensive question structure (is this authority a real expert here?) that resists it.\n\n**Modern catastrophic examples** include:\n\n**Theranos (founded 2003, collapsed 2018).** Elizabeth Holmes recruited a board with extraordinary authority signals — former Secretaries of State (Kissinger, Shultz), Stanford-affiliated medical figures, decorated military officers. Investors deferred. Walgreens, Safeway, the Department of Defense made commitments based on authority signals rather than scientific due diligence. The actual technology never worked. Patients received inaccurate medical results. Total losses: billions of dollars; loss of trust in lab innovation; criminal convictions for Holmes. The authority bias suppressed due diligence that would have caught the fraud.\n\n**FTX (founded 2019, collapsed November 2022).** Sam Bankman-Fried carried MIT credentials, prominent VC endorsements (Sequoia, Tiger Global, Paradigm), and political connections (largest political donor in many cycles). Investors and customers deferred. The collapse revealed $8 billion+ of customer funds had been misappropriated; the company had no functional risk management or auditing; the leadership team was making decisions in group chats with no oversight. The authority signals had suppressed the scrutiny that would have caught the fraud earlier. Many of the \"smart money\" investors (including major hedge funds and family offices) lost their entire investments.\n\nThe pattern: organizations that should have caught these frauds did not, because authority signals were strong and the structural countermeasures (independent due diligence, demanding evidence rather than credentials, applying skepticism to charismatic founders) were absent. Authority bias isn't a peripheral cognitive quirk; it's a load-bearing failure mode in modern finance and business.\n\nThe framework has been applied operationally in:\n\n**Medical decision-making.** Modern medicine has moved away from the \"white coat says so\" model toward shared decision-making, evidence-based protocols, and second-opinion practices. The structural changes are explicit authority-bias countermeasures.\n\n**Aviation crew resource management.** After several crashes attributed to captain authority overriding junior officer concerns (e.g., the 1977 Tenerife disaster), commercial aviation adopted Crew Resource Management protocols. These explicitly authorize junior officers to challenge captain decisions and require captains to actively solicit input.\n\n**Investment committees.** Sophisticated investment processes now include explicit \"kill the deal\" sessions where someone is assigned to argue against the investment, regardless of the partners' enthusiasm.\n\n**Organizational change.** Psychological safety research (Edmondson 1999) has demonstrated that teams where dissent is welcome substantially outperform teams where authority bias is unchecked. Google's Project Aristotle (2012-2015) found psychological safety to be the strongest predictor of team performance.\n\nThree operational lessons:\n\n**First, authority bias is automatic and survives explicit awareness.** Knowing about Milgram does not prevent you from deferring to authority. The cognitive mechanism is too deeply embedded. Structural countermeasures (premortems, devil's advocates, red teams, psychological safety) are required.\n\n**Second, distinguishing expertise from positional authority is the discipline.** A renowned investor's opinion on operations is positional, not expert. A charismatic CEO's claim about their technology is positional, not domain-expert if the CEO is not a technologist. The Cialdini two-question defense is the operational test.\n\n**Third, when you are the authority, you bear specific responsibility.** Your words and presence are systematically amplified, often beyond your intent. The structural counters (explicit dissent encouragement, separating suggestion from directive, asking explicitly for opposing views) compensate for the asymmetric weight. Operating as if your words have normal weight is operating with unrecognized power, and that power produces predictable errors.\n\nFile v1.0.6:examples/united-173-1978-cockpit-authority-gradient-and-crm.md\n\n# Method in Action: United 173 and the Cockpit Authority Gradient (1978 → 1981)\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nOn December 28, 1978, United Airlines Flight 173, a DC-8 inbound to Portland, Oregon, experienced a landing gear anomaly on approach — an abnormal indication and an unusual noise when the gear extended. The captain entered a holding pattern to troubleshoot and to prepare the cabin for a possible emergency landing. The airplane held for roughly an hour. During that hour, the fuel ran out. The DC-8 crash-landed in a wooded suburban area about six miles short of the airport. Ten people aboard died, including the flight engineer. Post-crash examination showed the landing gear had been down and functional the whole time.\n\n**S1 — Identify.** The authority: a highly experienced captain, backed by the era's cockpit culture in which the captain's judgment was close to unchallengeable. The question: when to stop troubleshooting and land. The pressure to defer: a steep authority gradient — the first officer and flight engineer were structurally subordinate, and the norm was to advise, never to countermand.\n\n**S2 — Expertise vs. positional.** The captain was a genuine domain expert as an aviator. But the specific question in play — remaining fuel versus remaining time — was squarely instrumented on the flight engineer's panel. The flight engineer had the better read on the binding constraint. Positional authority outweighed the crew member with the most relevant information.\n\n**S3 — Examine reasoning.** The captain's reasoning was locally sound: an unresolved gear indication justifies caution and cabin preparation. But it silently assumed time was abundant. The NTSB found that both junior crew members made advisory remarks about the deteriorating fuel state, yet neither converted those hints into an assertive, unambiguous challenge. The logical gap — \"we are preparing carefully\" versus \"we are about to run out of fuel\" — was filled by deference to the four stripes rather than by arithmetic.\n\n**S4 — Check for dysfunction.** Textbook pattern: dissent expressed as mitigated speech (soft advisories, indirect mentions of fuel figures), no escalation when the advisories were not acted on, and a captain absorbed in one problem while the crew's concern never registered as a command-level input. The NTSB's probable-cause finding cited the captain's failure to monitor and respond to the fuel state, and — contributing — the other crew members' failure either to grasp the criticality or to successfully communicate their concern to the captain.\n\n**S5 — Cialdini defense (retrospective).** Q1 — genuine expert on this specific question? On flying the DC-8, yes; on the fuel-time tradeoff at that moment, the flight engineer's data outranked the captain's attention. Q2 — does the reasoning hold under independent scrutiny? No: independent arithmetic on fuel flow contradicted the plan to keep holding. Either N → discount the authority signal and demand the evidence. The evidence was on the panel.\n\n**S6 — Structural counter.** This is why United 173 matters more than most crashes: the NTSB explicitly recommended crew-coordination training addressing exactly this failure mode, and in 1981 United Airlines instituted the industry's first Crew Resource Management program, building on NASA research into flight-deck resource management. CRM operationalizes the skill's structural counters — juniors are trained and authorized to challenge assertively with escalating phraseology, and captains are trained to explicitly solicit input and treat it as data, not insubordination. Awareness alone was never expected to fix the bias; the industry installed structure. CRM subsequently spread to airlines worldwide and into medicine, firefighting, and maritime operations.\n\nThe mapped steps:\n1. Identify: captain as near-absolute authority; question = keep holding vs. land now; steep gradient suppressing challenge\n2. Expertise vs. positional: captain expert in flying, but the binding constraint (fuel) was the flight engineer's instrumented domain\n3. Examine reasoning: cautious troubleshooting was sound only under an unstated abundant-time assumption; the gap was filled by rank\n4. Dysfunction check: mitigated speech, hints not acted on, no escalation — dissent structurally muted\n5. Cialdini defense: Q2 fails — independent fuel arithmetic contradicted the plan; authority signal should have been discounted\n6. Structural counter: NTSB recommendation → United's 1981 CRM program → industry-wide assertive-challenge and input-solicitation protocols\n\nPrimary source: National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The CRM aftermath is documented in Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.), *Crew Resource Management.* Academic Press.\n\nFile v1.0.6:skill-card.md\n\n## Description:\n\nDetect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support.\n\nThis skill is ready for commercial/non-commercial use.\n\n## Publisher:\n\n[deciqai](https://clawhub.ai/user/deciqai)\n\n### License/Terms of Use:\n\nMIT-0\n\n## Use Case:\n\nEmployees, external users, and developers use this skill to evaluate whether a person, institution, AI system, or executive claim is being over-weighted because of status rather than domain-specific expertise and independently checkable reasoning. It supports board, investment, medical, technical, organizational, negotiation, and AI-adoption decisions where authority signals may suppress dissent or evidence review.\n\n### Deployment Geography for Use:\n\nGlobal\n\n## Known Risks and Mitigations:\n\nRisk: The skill can be used in high-stakes domains such as medical, financial, organizational, and AI-adoption decisions where cited examples or authority assessments may be incomplete or stale.\n\nMitigation: Treat the output as decision-support guidance and independently verify load-bearing facts, citations, and domain expertise before relying on it.\n\nRisk: The artifact includes historical and business examples in prose, and evidence.security notes that some source claims are cited in prose rather than enforced by tooling.\n\nMitigation: Review the referenced source list and any case-specific claims before applying the analysis to a consequential decision.\n\nRisk: The skill may encourage discounting authority signals even when the authority is a verified expert on the exact question and the decision is time-critical.\n\nMitigation: Follow the artifact's stated boundary condition: when narrow domain expertise is verified and speed matters, calibrated deference can be appropriate.\n\n## Reference(s):\n\n- [ClawHub Skill Page](https://clawhub.ai/deciqai/skills/authority-bias)\n- [Authority Bias Source List](references/sources.md)\n- [Milgram, Cialdini, Theranos, and FTX Example](examples/milgram-1963-cialdini-systematization-theranos-ftx.md)\n- [United 173 and Cockpit Authority Gradient Example](examples/united-173-1978-cockpit-authority-gradient-and-crm.md)\n- [AI Authority Deference Example](examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md)\n- [deciqAI Authority Bias Page](https://www.deciqai.com/c/authority-bias)\n- [Machine-Readable Skill Metadata](https://www.deciqai.com/s/authority-bias.json)\n\n## Skill Output:\n\n**Output Type(s):** [text, markdown, guidance]\n\n**Output Format:** [Markdown-style analysis template with concise coaching questions and structured fields]\n\n**Output Parameters:** [1D]\n\n**Other Properties Related to Output:** [May pause for user input in coach mode before continuing the analysis.]\n\n## Skill Version(s):\n\n1.0.6 (source: server release evidence)\n\n## Ethical Considerations:\n\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.\n\nArchive v1.0.5: 7 files, 16115 bytes\n\nFiles: examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md (7518b), examples/milgram-1963-cialdini-systematization-theranos-ftx.md (5467b), examples/united-173-1978-cockpit-authority-gradient-and-crm.md (4990b), references/sources.md (2283b), skill-card.md (2662b), SKILL.md (8100b), _meta.json (133b)\n\nFile v1.0.5:SKILL.md\n\n---\nname: authority-bias\ndescription: \"Activate when: someone says 'should we trust the expert on this?', 'they're a professor/CEO/MD so they must be right', 'I don't want to push back on them', 'we deferred to the board and it blew up', or is evaluating a charismatic founder, assessing due diligence on a high-credential deal, or building a team where senior voices dominate.\nDo NOT activate when: the authority is a verified domain expert on the exact question at hand and the decision is time-critical; or when the user is asking a general persuasion question with no authority dynamic involved.\"\n---\n\n# Authority Bias\n\n## Overview\n\n**Authority bias** is the automatic tendency to comply with directives from authority figures — often overriding one's own judgment, ethics, or evidence. It operates beneath conscious deliberation: people substitute the authority's judgment for their own rather than weighing it.\n\nMilgram (1963) showed 65% of ordinary adults administered apparent 450V shocks when instructed by a lab-coated experimenter — against expert predictions of 1-2%. Cialdini (1984) systematized three authority triggers: titles, attire, and trappings — all of which fire regardless of actual expertise.\n\nKey distinction: **positional authority** (CEO, board chair, VC partner) vs. **domain expertise** (demonstrated track record on this specific question type). The bias treats them as identical. The skill does not.\n\nComposes with `social-proof`, `reciprocity`, `critical-thinking`, `signaling-games`, `dunning-kruger`.\n\n## When to Use\n\n- Boards, investment committees, due diligence on high-credential founders\n- Medical or technical decisions where expert opinion dominates discussion\n- Org decisions where senior voices suppress dissent; 360-feedback design\n- Negotiations where counterparty leverages credentials or position\n- Trusting a confident AI/LLM answer or an AI-lab leader's capability/timeline pronouncement because of status rather than verified evidence (AI adoption, AI hype)\n\n**Not when:** authority is verified domain expert on the exact question and speed matters.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** concrete case → run The Process directly.\n- **Coach mode:** unfamiliar or no case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: before deferring, ask whether they're a genuine expert on this specific question and whether their reasoning holds up independently.\n2. Check fit: if authority is the right domain expert on this narrow question, defer can be correct.\n3. Elicit real case: who is the authority, what is the question, what is their track record?\n> **[WAIT — do not advance until user responds]**\n4. Run S1-S5 one step at a time with their input.\n> **[WAIT — do not advance until user responds]**\n5. Close: name the insight — calibrated weighting + structural counters if relevant.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**S1 — Identify:** Authority (person/institution) · question/decision · authority signals (title, attire, credentials) · pressure to defer.\n\n**S2 — Expertise vs. positional:** Actual track record on this question type? Positional (CEO, chair) or domain expertise (years operating here)? Would a peer expert agree?\n\n**S3 — Examine reasoning:** Is the reasoning sound independent of who's stating it? Are logical gaps being filled by authority signals?\n\n**S4 — Check for dysfunction:** Is dissent suppressed? Are decisions justified by \"X said so\"? Is positional authority conflated with domain expertise?\n\n**S5 — Cialdini defense:** (Q1) Genuine expert on this specific question? (Q2) Reasoning holds under independent scrutiny? Both Y → appropriate weight. Either N → discount signal, demand evidence.\n\n**S6 — Structural counter (if you are the authority):** Explicitly invite dissent (\"give me 3 reasons against\") · separate suggestions from directives · install red team / premortem.\n\n## Output: Authority-Bias Analysis\n```\nAuthority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:\n```\n\n*→ Method in Action: [Milgram 1963 + Cialdini Systematization + Theranos/FTX](examples/milgram-1963-cialdini-systematization-theranos-ftx.md) · [United 173 and the Cockpit Authority Gradient](examples/united-173-1978-cockpit-authority-gradient-and-crm.md)*\n*→ 2026 lens: [Deferring to AI Authority — Confident LLMs and Lab-Leader Pronouncements (2024–2026)](examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md)*\n\n## Pack: Authority-Bias Application Patterns\n\n| Situation | Defensive question |\n|---|---|\n| Investment in startup (CEO / VC brand) | Have they demonstrated this specific business model works? |\n| Board member's opinion | Is this question actually in their domain expertise? |\n| Medical recommendation | What's the evidence base, independent of the white coat? |\n| Senior engineer's design decision | Has this approach worked in our specific context before? |\n| Founder's product vision | Are users requesting this, or is it founder preference? |\n\n## Applying It Well\n\n- Awareness doesn't prevent deference — structural countermeasures are required.\n- When you are the authority, your words are amplified beyond intent; operate accordingly.\n- Cialdini defense: (a) genuine expert on this question? (b) reasoning holds independently?\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"They're smart; we should listen\" | Smart in general doesn't mean smart on this specific question. Apply the domain test. |\n| [D] \"Everyone respects them\" | Cialdini's social proof reinforcing authority. Both biases compound. Separate them. |\n| [D] \"They have a great track record\" | In what specifically? Apply the question scope check. |\n| [D] \"It would be rude to push back\" | Compliance despite better judgment is the bias operating. Rudeness is not the actual cost. |\n| [D] \"I'm not in a position to disagree\" | If you have contradicting evidence or reasoning, you are. The framing is authority bias self-reinforcing. |\n| [D] \"If they're wrong, it'll show in due diligence\" | Theranos and FTX survived layers of due diligence by leveraging authority to suppress scrutiny. |\n| [D] \"I'm too junior\" | Junior people often have unfiltered observations. Suppressing them is where structural failures occur. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\n- Decision made primarily because an authority said so, not because the reasoning holds\n- Dissent suppressed — junior people silent, mid-managers deferring without pushback\n- Due diligence shortened due to authority signals; credentials conflated with domain expertise\n- No red team / premortem / devil's advocate on a high-stakes call\n\n## Verification\n\n- [ ] Authority's track record on this specific question examined\n- [ ] Positional authority distinguished from domain expertise\n- [ ] Reasoning independently examined; at least one dissenting expert sought\n- [ ] Cialdini two-question defense applied\n- [ ] If org: psychological safety mechanisms in place\n- [ ] If you are the authority: structural dissent-encouragement operational\n- [ ] Decision documented with reasoning, not just authority attribution\n\n*Part of **deciqAI Knowledge Skills** — 223 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/authority-bias** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.5:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"authority-bias\",\n  \"version\": \"1.0.5\",\n  \"publishedAt\": 1783679048661\n}\n\nFile v1.0.5:references/sources.md\n\n# Sources — authority-bias\n\n> *Primary sources for the [authority-bias](../SKILL.md) skill.*\n\n- Milgram, S. (1963). \"Behavioral Study of Obedience.\" *Journal of Abnormal and Social Psychology*, 67(4), 371-378. The foundational experiment.\n- Milgram, S. (1974). *Obedience to Authority: An Experimental View.* Harper & Row. ISBN 978-0061765216. The comprehensive treatment.\n- Cialdini, R. B. (1984/2006). *Influence: The Psychology of Persuasion.* HarperBusiness. ISBN 978-0061241895. Ch. 6.\n- Burger, J. M. (2009). \"Replicating Milgram: Would People Still Obey Today?\" *American Psychologist*, 64(1), 1-11. Modern replication.\n- Edmondson, A. C. (1999). \"Psychological Safety and Learning Behavior in Work Teams.\" *Administrative Science Quarterly*, 44(2), 350-383.\n- Janis, I. L. (1972). *Victims of Groupthink.* Houghton Mifflin. ISBN 978-0395140024. Group dysfunction and authority dynamics.\n- Carreyrou, J. (2018). *Bad Blood: Secrets and Lies in a Silicon Valley Startup.* Knopf. ISBN 978-1524731656. The Theranos case.\n- Lewis, M. (2023). *Going Infinite: The Rise and Fall of a New Tycoon.* W. W. Norton. ISBN 978-1324074335. The FTX case.\n- Heath, C. & Heath, D. (2013). *Decisive: How to Make Better Choices in Life and Work.* Crown. Includes authority-bias countermeasures.\n- National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The cockpit authority gradient case that led to Crew Resource Management.\n- Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.) (2010). *Crew Resource Management.* Academic Press. CRM as a structural counter to cockpit authority bias.\n- *Mata v. Avianca, Inc.*, No. 22-cv-1461 (S.D.N.Y. 2023). U.S. District Court sanctions against attorneys who submitted a brief with fabricated case citations generated by ChatGPT — canonical example of deferring to confident-but-wrong AI output. Widely reported (The New York Times, Reuters, 2023).\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. On confidence miscalibration and the poor track record of expert forecasting — the basis for discounting fluent AI answers and famous-founder timeline claims by status.\n\nFile v1.0.5:examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md\n\n# Method in Action: Deferring to AI Authority — Confident LLMs and Lab-Leader Pronouncements (2024–2026)\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nBy 2024–2026, two new authority signals became routine inputs to real decisions. First, large language models answer in fluent, confident, well-formatted prose — a rhetorical register that reads as expertise regardless of whether the underlying answer is correct. Second, the leaders of frontier AI labs make widely-quoted pronouncements about capability and timelines (when models will reach a given benchmark, when a class of jobs is automated, when \"AGI\" arrives), and those pronouncements are treated as forecasts from the highest possible authority. In both cases the risk is the same: weighting the source by its **status** — the polished output, the founder's fame, the lab's brand — rather than by its **demonstrated track record on the specific question being asked**. This is authority bias wearing a 2020s costume.\n\n**S1 — Identify.** Two overlapping authorities. (a) *The model itself*: an LLM produces an answer to a factual, legal, medical, or financial question. Authority signals — fluency, confidence, citations-shaped text, a trusted brand name on the product. Pressure to defer: it's fast, it sounds sure, and checking is work. (b) *The lab leader*: a well-known founder/CEO states a capability or timeline claim in an interview or post. Signals — title, fame, proximity to the technology, prior correct-sounding calls. Pressure to defer: they build the thing, so surely they know. The question in each case is narrow: *is this specific answer correct?* / *is this specific timeline claim well-supported?*\n\n**S2 — Expertise vs. positional.** Apply the skill's core distinction. An LLM has no track record on *your* specific question; it has an aggregate tendency to produce plausible text, and its confidence is not calibrated to its accuracy — models are widely documented to state wrong answers (including fabricated citations and cases) in exactly the same authoritative tone as correct ones. Fluency is a *positional* signal (it looks authoritative), not *domain expertise* on the instant question. For the lab leader: they are genuine experts at *building and running an AI lab*. But a claim like \"capability X arrives by year Y\" is a forecast about a complex, partly external system — and forecasters (including insiders) have a poor documented record on technology timelines. Worse, the lab leader is not a disinterested forecaster: bullish public timelines can help recruiting, fundraising, and valuation. Positional authority and commercial incentive are being read as neutral domain expertise.\n\n**S3 — Examine reasoning.** Strip the source and look at the argument. For the LLM answer: is there a verifiable chain — a real citation you can open, a computation you can redo, a source you can check — or is the confidence doing the work the evidence should be doing? The failure mode is treating \"the model asserted it fluently\" as \"the model verified it.\" A now-canonical illustration: in 2023 a U.S. federal court sanctioned lawyers who filed a brief citing court cases that a chatbot had fabricated; the confident, correctly-formatted citations were entirely invented. For the timeline claim: is it backed by an argument with checkable premises (measured trend, defined benchmark, stated assumptions), or is it an unfalsifiable vibe anchored to a round-numbered year? Note also that prominent predictions of near-term milestones have repeatedly slipped — the gap between \"confidently stated by an authority\" and \"borne out\" is exactly the space authority bias hides in.\n\n**S4 — Check for dysfunction.** In organizations adopting AI in 2024–2026, watch for: decisions justified by \"the AI said so\" with no human verification of the load-bearing facts; strategy decks whose headline assumption is a single lab leader's quoted timeline, unchallenged because of who said it; junior analysts who spotted a hallucinated figure but stayed quiet because the tool (or the executive citing it) carried more perceived authority than their observation. This is the same mitigated-dissent pattern as any authority-gradient failure — only now one of the \"authorities\" is a text generator with no accountability, and the other is an interested party.\n\n**S5 — Cialdini defense.** (Q1) Genuine expert on this *specific* question? For the LLM: no — it has no verified competence on your instant fact; treat its output as a *draft to be checked*, not a ruling. For the lab leader: expert on building labs, *not* a calibrated oracle on external timelines, and conflicted. (Q2) Does the reasoning hold under independent scrutiny? Only if the citation opens, the number reconciles, the benchmark is defined and measured. Either answer N → **discount the authority signal and demand the evidence.** In both cases at least one answer is N by default, so the default posture is: verify before you rely.\n\n**S6 — Structural counter (if you are the authority, or the buyer of the authority).** Install structure rather than relying on awareness. For LLM output: require a human-in-the-loop verification step for any AI-asserted fact that is load-bearing (every case citation opened, every figure traced to a primary source), and log the source, not just the AI attribution. For timeline/capability claims: separate \"a founder said it\" from \"it is established\"; demand the underlying benchmark, trend, and assumptions; seek an independent, disinterested expert; and run a premortem (\"if this timeline is wrong by 3x, what breaks?\"). Treat confident AI text and famous-founder forecasts identically to any other authority signal — as an input to be weighted by track record on *this* question, never as a substitute for the evidence.\n\nThe mapped steps:\n1. Identify: two authorities — the confident LLM answer and the lab-leader pronouncement; narrow question = is *this* answer / *this* timeline correct\n2. Expertise vs. positional: fluency and fame are positional signals; neither is calibrated domain expertise on the instant question, and the leader is commercially interested\n3. Examine reasoning: does a citation open / a number reconcile / a benchmark exist — or is confidence filling the evidentiary gap (fabricated-citation sanctions as the canonical failure)\n4. Dysfunction check: \"the AI said so\" with no verification; strategy anchored to one quoted timeline; suppressed junior observations\n5. Cialdini defense: Q1 and Q2 both default to N — discount the signal, demand the evidence, verify before relying\n6. Structural counter: human-in-the-loop fact verification, source logging, independent disinterested experts, timeline premortems\n\n*Sources: Kahneman, D. (2011). Thinking, Fast and Slow, Farrar, Straus and Giroux — on the miscalibration of confidence and the poor track record of expert forecasting. Mata v. Avianca, Inc., No. 22-cv-1461 (S.D.N.Y. 2023) — U.S. District Court sanctions against attorneys who submitted a brief containing fabricated case citations generated by ChatGPT (widely reported, e.g., by The New York Times and Reuters, 2023). Cialdini, R. B. (1984/2006). Influence: The Psychology of Persuasion, HarperBusiness, Ch. 6 — the authority principle and its title/attire/trappings triggers. General LLM hallucination and confidence-miscalibration behavior is documented across model-provider system/model cards and the academic literature as of early 2026.*\n\nFile v1.0.5:examples/milgram-1963-cialdini-systematization-theranos-ftx.md\n\n# Method in Action: Milgram 1963 + Cialdini Systematization + Theranos/FTX\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nThe **Milgram 1963 experiment** is the foundational empirical case. The contrast between psychologist prediction (1-2% compliance) and actual result (65%) is the canonical demonstration of how little of human behavior is governed by the rational, ethical, individual judgment that we believe we possess and that observers believe we possess. Under authority pressure, most ordinary adults will administer apparent severe harm — not because they are bad people, but because authority compliance is a deeply embedded social instinct.\n\nMilgram's 1974 follow-up book provided the systematic treatment, including the 18+ variations that mapped the conditions of authority compliance. The pattern: legitimacy of authority, presence of authority, absence of dissenting peers, distance from the victim all increase compliance. The reverse conditions decrease it. The framework has been validated in cross-cultural replications, including Burger's 2009 partial replication.\n\n**Robert Cialdini's 1984 systematization** placed authority among the six universal weapons of influence. Cialdini's contribution was operational: identifying the specific signals (titles, attire, trappings) that trigger compliance and the defensive question structure (is this authority a real expert here?) that resists it.\n\n**Modern catastrophic examples** include:\n\n**Theranos (founded 2003, collapsed 2018).** Elizabeth Holmes recruited a board with extraordinary authority signals — former Secretaries of State (Kissinger, Shultz), Stanford-affiliated medical figures, decorated military officers. Investors deferred. Walgreens, Safeway, the Department of Defense made commitments based on authority signals rather than scientific due diligence. The actual technology never worked. Patients received inaccurate medical results. Total losses: billions of dollars; loss of trust in lab innovation; criminal convictions for Holmes. The authority bias suppressed due diligence that would have caught the fraud.\n\n**FTX (founded 2019, collapsed November 2022).** Sam Bankman-Fried carried MIT credentials, prominent VC endorsements (Sequoia, Tiger Global, Paradigm), and political connections (largest political donor in many cycles). Investors and customers deferred. The collapse revealed $8 billion+ of customer funds had been misappropriated; the company had no functional risk management or auditing; the leadership team was making decisions in group chats with no oversight. The authority signals had suppressed the scrutiny that would have caught the fraud earlier. Many of the \"smart money\" investors (including major hedge funds and family offices) lost their entire investments.\n\nThe pattern: organizations that should have caught these frauds did not, because authority signals were strong and the structural countermeasures (independent due diligence, demanding evidence rather than credentials, applying skepticism to charismatic founders) were absent. Authority bias isn't a peripheral cognitive quirk; it's a load-bearing failure mode in modern finance and business.\n\nThe framework has been applied operationally in:\n\n**Medical decision-making.** Modern medicine has moved away from the \"white coat says so\" model toward shared decision-making, evidence-based protocols, and second-opinion practices. The structural changes are explicit authority-bias countermeasures.\n\n**Aviation crew resource management.** After several crashes attributed to captain authority overriding junior officer concerns (e.g., the 1977 Tenerife disaster), commercial aviation adopted Crew Resource Management protocols. These explicitly authorize junior officers to challenge captain decisions and require captains to actively solicit input.\n\n**Investment committees.** Sophisticated investment processes now include explicit \"kill the deal\" sessions where someone is assigned to argue against the investment, regardless of the partners' enthusiasm.\n\n**Organizational change.** Psychological safety research (Edmondson 1999) has demonstrated that teams where dissent is welcome substantially outperform teams where authority bias is unchecked. Google's Project Aristotle (2012-2015) found psychological safety to be the strongest predictor of team performance.\n\nThree operational lessons:\n\n**First, authority bias is automatic and survives explicit awareness.** Knowing about Milgram does not prevent you from deferring to authority. The cognitive mechanism is too deeply embedded. Structural countermeasures (premortems, devil's advocates, red teams, psychological safety) are required.\n\n**Second, distinguishing expertise from positional authority is the discipline.** A renowned investor's opinion on operations is positional, not expert. A charismatic CEO's claim about their technology is positional, not domain-expert if the CEO is not a technologist. The Cialdini two-question defense is the operational test.\n\n**Third, when you are the authority, you bear specific responsibility.** Your words and presence are systematically amplified, often beyond your intent. The structural counters (explicit dissent encouragement, separating suggestion from directive, asking explicitly for opposing views) compensate for the asymmetric weight. Operating as if your words have normal weight is operating with unrecognized power, and that power produces predictable errors.\n\nFile v1.0.5:examples/united-173-1978-cockpit-authority-gradient-and-crm.md\n\n# Method in Action: United 173 and the Cockpit Authority Gradient (1978 → 1981)\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nOn December 28, 1978, United Airlines Flight 173, a DC-8 inbound to Portland, Oregon, experienced a landing gear anomaly on approach — an abnormal indication and an unusual noise when the gear extended. The captain entered a holding pattern to troubleshoot and to prepare the cabin for a possible emergency landing. The airplane held for roughly an hour. During that hour, the fuel ran out. The DC-8 crash-landed in a wooded suburban area about six miles short of the airport. Ten people aboard died, including the flight engineer. Post-crash examination showed the landing gear had been down and functional the whole time.\n\n**S1 — Identify.** The authority: a highly experienced captain, backed by the era's cockpit culture in which the captain's judgment was close to unchallengeable. The question: when to stop troubleshooting and land. The pressure to defer: a steep authority gradient — the first officer and flight engineer were structurally subordinate, and the norm was to advise, never to countermand.\n\n**S2 — Expertise vs. positional.** The captain was a genuine domain expert as an aviator. But the specific question in play — remaining fuel versus remaining time — was squarely instrumented on the flight engineer's panel. The flight engineer had the better read on the binding constraint. Positional authority outweighed the crew member with the most relevant information.\n\n**S3 — Examine reasoning.** The captain's reasoning was locally sound: an unresolved gear indication justifies caution and cabin preparation. But it silently assumed time was abundant. The NTSB found that both junior crew members made advisory remarks about the deteriorating fuel state, yet neither converted those hints into an assertive, unambiguous challenge. The logical gap — \"we are preparing carefully\" versus \"we are about to run out of fuel\" — was filled by deference to the four stripes rather than by arithmetic.\n\n**S4 — Check for dysfunction.** Textbook pattern: dissent expressed as mitigated speech (soft advisories, indirect mentions of fuel figures), no escalation when the advisories were not acted on, and a captain absorbed in one problem while the crew's concern never registered as a command-level input. The NTSB's probable-cause finding cited the captain's failure to monitor and respond to the fuel state, and — contributing — the other crew members' failure either to grasp the criticality or to successfully communicate their concern to the captain.\n\n**S5 — Cialdini defense (retrospective).** Q1 — genuine expert on this specific question? On flying the DC-8, yes; on the fuel-time tradeoff at that moment, the flight engineer's data outranked the captain's attention. Q2 — does the reasoning hold under independent scrutiny? No: independent arithmetic on fuel flow contradicted the plan to keep holding. Either N → discount the authority signal and demand the evidence. The evidence was on the panel.\n\n**S6 — Structural counter.** This is why United 173 matters more than most crashes: the NTSB explicitly recommended crew-coordination training addressing exactly this failure mode, and in 1981 United Airlines instituted the industry's first Crew Resource Management program, building on NASA research into flight-deck resource management. CRM operationalizes the skill's structural counters — juniors are trained and authorized to challenge assertively with escalating phraseology, and captains are trained to explicitly solicit input and treat it as data, not insubordination. Awareness alone was never expected to fix the bias; the industry installed structure. CRM subsequently spread to airlines worldwide and into medicine, firefighting, and maritime operations.\n\nThe mapped steps:\n1. Identify: captain as near-absolute authority; question = keep holding vs. land now; steep gradient suppressing challenge\n2. Expertise vs. positional: captain expert in flying, but the binding constraint (fuel) was the flight engineer's instrumented domain\n3. Examine reasoning: cautious troubleshooting was sound only under an unstated abundant-time assumption; the gap was filled by rank\n4. Dysfunction check: mitigated speech, hints not acted on, no escalation — dissent structurally muted\n5. Cialdini defense: Q2 fails — independent fuel arithmetic contradicted the plan; authority signal should have been discounted\n6. Structural counter: NTSB recommendation → United's 1981 CRM program → industry-wide assertive-challenge and input-solicitation protocols\n\nPrimary source: National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The CRM aftermath is documented in Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.), *Crew Resource Management.* Academic Press.\n\nFile v1.0.5:skill-card.md\n\n## Description: <br>\nDetect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and developers use this skill to analyze decisions where credentials, role, charisma, or institutional status may be substituting for evidence. It guides users through identifying the authority signal, checking domain-specific expertise, testing the reasoning independently, and installing structural counters for dissent. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Users may treat historical, business, legal, medical, financial, or safety examples as a substitute for professional advice. <br>\nMitigation: Use the skill as decision-support guidance and verify load-bearing decisions with qualified experts, primary evidence, and applicable organizational review. <br>\nRisk: The skill may overcorrect against valid expertise in time-critical situations. <br>\nMitigation: Apply the skill's stated exception: appropriate deference can be warranted when the authority is a verified domain expert on the exact question and speed matters. <br>\n\n\n## Reference(s): <br>\n- [ClawHub Authority Bias Skill Page](https://clawhub.ai/deciqai/skills/authority-bias) <br>\n- [Sources - authority-bias](artifact/references/sources.md) <br>\n- [Milgram 1963 + Cialdini Systematization + Theranos/FTX](artifact/examples/milgram-1963-cialdini-systematization-theranos-ftx.md) <br>\n- [United 173 and the Cockpit Authority Gradient](artifact/examples/united-173-1978-cockpit-authority-gradient-and-crm.md) <br>\n- [Deferring to AI Authority](artifact/examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [guidance, markdown, text] <br>\n**Output Format:** [Markdown structured as an authority-bias analysis template and coaching questions] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [Plain-text decision-support output; no executable code or hidden access requests.] <br>\n\n## Skill Version(s): <br>\n1.0.5 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.4: 6 files, 11754 bytes\n\nFiles: examples/milgram-1963-cialdini-systematization-theranos-ftx.md (5467b), examples/united-173-1978-cockpit-authority-gradient-and-crm.md (4990b), references/sources.md (1711b), skill-card.md (2349b), SKILL.md (7751b), _meta.json (133b)\n\nFile v1.0.4:SKILL.md\n\n---\nname: authority-bias\ndescription: \"Activate when: someone says 'should we trust the expert on this?', 'they're a professor/CEO/MD so they must be right', 'I don't want to push back on them', 'we deferred to the board and it blew up', or is evaluating a charismatic founder, assessing due diligence on a high-credential deal, or building a team where senior voices dominate.\nDo NOT activate when: the authority is a verified domain expert on the exact question at hand and the decision is time-critical; or when the user is asking a general persuasion question with no authority dynamic involved.\"\n---\n\n# Authority Bias\n\n## Overview\n\n**Authority bias** is the automatic tendency to comply with directives from authority figures — often overriding one's own judgment, ethics, or evidence. It operates beneath conscious deliberation: people substitute the authority's judgment for their own rather than weighing it.\n\nMilgram (1963) showed 65% of ordinary adults administered apparent 450V shocks when instructed by a lab-coated experimenter — against expert predictions of 1-2%. Cialdini (1984) systematized three authority triggers: titles, attire, and trappings — all of which fire regardless of actual expertise.\n\nKey distinction: **positional authority** (CEO, board chair, VC partner) vs. **domain expertise** (demonstrated track record on this specific question type). The bias treats them as identical. The skill does not.\n\nComposes with `social-proof`, `reciprocity`, `critical-thinking`, `signaling-games`, `dunning-kruger`.\n\n## When to Use\n\n- Boards, investment committees, due diligence on high-credential founders\n- Medical or technical decisions where expert opinion dominates discussion\n- Org decisions where senior voices suppress dissent; 360-feedback design\n- Negotiations where counterparty leverages credentials or position\n\n**Not when:** authority is verified domain expert on the exact question and speed matters.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** concrete case → run The Process directly.\n- **Coach mode:** unfamiliar or no case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: before deferring, ask whether they're a genuine expert on this specific question and whether their reasoning holds up independently.\n2. Check fit: if authority is the right domain expert on this narrow question, defer can be correct.\n3. Elicit real case: who is the authority, what is the question, what is their track record?\n> **[WAIT — do not advance until user responds]**\n4. Run S1-S5 one step at a time with their input.\n> **[WAIT — do not advance until user responds]**\n5. Close: name the insight — calibrated weighting + structural counters if relevant.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**S1 — Identify:** Authority (person/institution) · question/decision · authority signals (title, attire, credentials) · pressure to defer.\n\n**S2 — Expertise vs. positional:** Actual track record on this question type? Positional (CEO, chair) or domain expertise (years operating here)? Would a peer expert agree?\n\n**S3 — Examine reasoning:** Is the reasoning sound independent of who's stating it? Are logical gaps being filled by authority signals?\n\n**S4 — Check for dysfunction:** Is dissent suppressed? Are decisions justified by \"X said so\"? Is positional authority conflated with domain expertise?\n\n**S5 — Cialdini defense:** (Q1) Genuine expert on this specific question? (Q2) Reasoning holds under independent scrutiny? Both Y → appropriate weight. Either N → discount signal, demand evidence.\n\n**S6 — Structural counter (if you are the authority):** Explicitly invite dissent (\"give me 3 reasons against\") · separate suggestions from directives · install red team / premortem.\n\n## Output: Authority-Bias Analysis\n```\nAuthority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:\n```\n\n*→ Method in Action: [Milgram 1963 + Cialdini Systematization + Theranos/FTX](examples/milgram-1963-cialdini-systematization-theranos-ftx.md) · [United 173 and the Cockpit Authority Gradient](examples/united-173-1978-cockpit-authority-gradient-and-crm.md)*\n\n## Pack: Authority-Bias Application Patterns\n\n| Situation | Defensive question |\n|---|---|\n| Investment in startup (CEO / VC brand) | Have they demonstrated this specific business model works? |\n| Board member's opinion | Is this question actually in their domain expertise? |\n| Medical recommendation | What's the evidence base, independent of the white coat? |\n| Senior engineer's design decision | Has this approach worked in our specific context before? |\n| Founder's product vision | Are users requesting this, or is it founder preference? |\n\n## Applying It Well\n\n- Awareness doesn't prevent deference — structural countermeasures are required.\n- When you are the authority, your words are amplified beyond intent; operate accordingly.\n- Cialdini defense: (a) genuine expert on this question? (b) reasoning holds independently?\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"They're smart; we should listen\" | Smart in general doesn't mean smart on this specific question. Apply the domain test. |\n| [D] \"Everyone respects them\" | Cialdini's social proof reinforcing authority. Both biases compound. Separate them. |\n| [D] \"They have a great track record\" | In what specifically? Apply the question scope check. |\n| [D] \"It would be rude to push back\" | Compliance despite better judgment is the bias operating. Rudeness is not the actual cost. |\n| [D] \"I'm not in a position to disagree\" | If you have contradicting evidence or reasoning, you are. The framing is authority bias self-reinforcing. |\n| [D] \"If they're wrong, it'll show in due diligence\" | Theranos and FTX survived layers of due diligence by leveraging authority to suppress scrutiny. |\n| [D] \"I'm too junior\" | Junior people often have unfiltered observations. Suppressing them is where structural failures occur. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\n- Decision made primarily because an authority said so, not because the reasoning holds\n- Dissent suppressed — junior people silent, mid-managers deferring without pushback\n- Due diligence shortened due to authority signals; credentials conflated with domain expertise\n- No red team / premortem / devil's advocate on a high-stakes call\n\n## Verification\n\n- [ ] Authority's track record on this specific question examined\n- [ ] Positional authority distinguished from domain expertise\n- [ ] Reasoning independently examined; at least one dissenting expert sought\n- [ ] Cialdini two-question defense applied\n- [ ] If org: psychological safety mechanisms in place\n- [ ] If you are the authority: structural dissent-encouragement operational\n- [ ] Decision documented with reasoning, not just authority attribution\n\n*Part of **deciqAI Knowledge Skills** — 164 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/c/authority-bias** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.4:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"authority-bias\",\n  \"version\": \"1.0.4\",\n  \"publishedAt\": 1783508032857\n}\n\nFile v1.0.4:references/sources.md\n\n# Sources — authority-bias\n\n> *Primary sources for the [authority-bias](../SKILL.md) skill.*\n\n- Milgram, S. (1963). \"Behavioral Study of Obedience.\" *Journal of Abnormal and Social Psychology*, 67(4), 371-378. The foundational experiment.\n- Milgram, S. (1974). *Obedience to Authority: An Experimental View.* Harper & Row. ISBN 978-0061765216. The comprehensive treatment.\n- Cialdini, R. B. (1984/2006). *Influence: The Psychology of Persuasion.* HarperBusiness. ISBN 978-0061241895. Ch. 6.\n- Burger, J. M. (2009). \"Replicating Milgram: Would People Still Obey Today?\" *American Psychologist*, 64(1), 1-11. Modern replication.\n- Edmondson, A. C. (1999). \"Psychological Safety and Learning Behavior in Work Teams.\" *Administrative Science Quarterly*, 44(2), 350-383.\n- Janis, I. L. (1972). *Victims of Groupthink.* Houghton Mifflin. ISBN 978-0395140024. Group dysfunction and authority dynamics.\n- Carreyrou, J. (2018). *Bad Blood: Secrets and Lies in a Silicon Valley Startup.* Knopf. ISBN 978-1524731656. The Theranos case.\n- Lewis, M. (2023). *Going Infinite: The Rise and Fall of a New Tycoon.* W. W. Norton. ISBN 978-1324074335. The FTX case.\n- Heath, C. & Heath, D. (2013). *Decisive: How to Make Better Choices in Life and Work.* Crown. Includes authority-bias countermeasures.\n- National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The cockpit authority gradient case that led to Crew Resource Management.\n- Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.) (2010). *Crew Resource Management.* Academic Press. CRM as a structural counter to cockpit authority bias.\n\nFile v1.0.4:examples/milgram-1963-cialdini-systematization-theranos-ftx.md\n\n# Method in Action: Milgram 1963 + Cialdini Systematization + Theranos/FTX\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nThe **Milgram 1963 experiment** is the foundational empirical case. The contrast between psychologist prediction (1-2% compliance) and actual result (65%) is the canonical demonstration of how little of human behavior is governed by the rational, ethical, individual judgment that we believe we possess and that observers believe we possess. Under authority pressure, most ordinary adults will administer apparent severe harm — not because they are bad people, but because authority compliance is a deeply embedded social instinct.\n\nMilgram's 1974 follow-up book provided the systematic treatment, including the 18+ variations that mapped the conditions of authority compliance. The pattern: legitimacy of authority, presence of authority, absence of dissenting peers, distance from the victim all increase compliance. The reverse conditions decrease it. The framework has been validated in cross-cultural replications, including Burger's 2009 partial replication.\n\n**Robert Cialdini's 1984 systematization** placed authority among the six universal weapons of influence. Cialdini's contribution was operational: identifying the specific signals (titles, attire, trappings) that trigger compliance and the defensive question structure (is this authority a real expert here?) that resists it.\n\n**Modern catastrophic examples** include:\n\n**Theranos (founded 2003, collapsed 2018).** Elizabeth Holmes recruited a board with extraordinary authority signals — former Secretaries of State (Kissinger, Shultz), Stanford-affiliated medical figures, decorated military officers. Investors deferred. Walgreens, Safeway, the Department of Defense made commitments based on authority signals rather than scientific due diligence. The actual technology never worked. Patients received inaccurate medical results. Total losses: billions of dollars; loss of trust in lab innovation; criminal convictions for Holmes. The authority bias suppressed due diligence that would have caught the fraud.\n\n**FTX (founded 2019, collapsed November 2022).** Sam Bankman-Fried carried MIT credentials, prominent VC endorsements (Sequoia, Tiger Global, Paradigm), and political connections (largest political donor in many cycles). Investors and customers deferred. The collapse revealed $8 billion+ of customer funds had been misappropriated; the company had no functional risk management or auditing; the leadership team was making decisions in group chats with no oversight. The authority signals had suppressed the scrutiny that would have caught the fraud earlier. Many of the \"smart money\" investors (including major hedge funds and family offices) lost their entire investments.\n\nThe pattern: organizations that should have caught these frauds did not, because authority signals were strong and the structural countermeasures (independent due diligence, demanding evidence rather than credentials, applying skepticism to charismatic founders) were absent. Authority bias isn't a peripheral cognitive quirk; it's a load-bearing failure mode in modern finance and business.\n\nThe framework has been applied operationally in:\n\n**Medical decision-making.** Modern medicine has moved away from the \"white coat says so\" model toward shared decision-making, evidence-based protocols, and second-opinion practices. The structural changes are explicit authority-bias countermeasures.\n\n**Aviation crew resource management.** After several crashes attributed to captain authority overriding junior officer concerns (e.g., the 1977 Tenerife disaster), commercial aviation adopted Crew Resource Management protocols. These explicitly authorize junior officers to challenge captain decisions and require captains to actively solicit input.\n\n**Investment committees.** Sophisticated investment processes now include explicit \"kill the deal\" sessions where someone is assigned to argue against the investment, regardless of the partners' enthusiasm.\n\n**Organizational change.** Psychological safety research (Edmondson 1999) has demonstrated that teams where dissent is welcome substantially outperform teams where authority bias is unchecked. Google's Project Aristotle (2012-2015) found psychological safety to be the strongest predictor of team performance.\n\nThree operational lessons:\n\n**First, authority bias is automatic and survives explicit awareness.** Knowing about Milgram does not prevent you from deferring to authority. The cognitive mechanism is too deeply embedded. Structural countermeasures (premortems, devil's advocates, red teams, psychological safety) are required.\n\n**Second, distinguishing expertise from positional authority is the discipline.** A renowned investor's opinion on operations is positional, not expert. A charismatic CEO's claim about their technology is positional, not domain-expert if the CEO is not a technologist. The Cialdini two-question defense is the operational test.\n\n**Third, when you are the authority, you bear specific responsibility.** Your words and presence are systematically amplified, often beyond your intent. The structural counters (explicit dissent encouragement, separating suggestion from directive, asking explicitly for opposing views) compensate for the asymmetric weight. Operating as if your words have normal weight is operating with unrecognized power, and that power produces predictable errors.\n\nFile v1.0.4:examples/united-173-1978-cockpit-authority-gradient-and-crm.md\n\n# Method in Action: United 173 and the Cockpit Authority Gradient (1978 → 1981)\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nOn December 28, 1978, United Airlines Flight 173, a DC-8 inbound to Portland, Oregon, experienced a landing gear anomaly on approach — an abnormal indication and an unusual noise when the gear extended. The captain entered a holding pattern to troubleshoot and to prepare the cabin for a possible emergency landing. The airplane held for roughly an hour. During that hour, the fuel ran out. The DC-8 crash-landed in a wooded suburban area about six miles short of the airport. Ten people aboard died, including the flight engineer. Post-crash examination showed the landing gear had been down and functional the whole time.\n\n**S1 — Identify.** The authority: a highly experienced captain, backed by the era's cockpit culture in which the captain's judgment was close to unchallengeable. The question: when to stop troubleshooting and land. The pressure to defer: a steep authority gradient — the first officer and flight engineer were structurally subordinate, and the norm was to advise, never to countermand.\n\n**S2 — Expertise vs. positional.** The captain was a genuine domain expert as an aviator. But the specific question in play — remaining fuel versus remaining time — was squarely instrumented on the flight engineer's panel. The flight engineer had the better read on the binding constraint. Positional authority outweighed the crew member with the most relevant information.\n\n**S3 — Examine reasoning.** The captain's reasoning was locally sound: an unresolved gear indication justifies caution and cabin preparation. But it silently assumed time was abundant. The NTSB found that both junior crew members made advisory remarks about the deteriorating fuel state, yet neither converted those hints into an assertive, unambiguous challenge. The logical gap — \"we are preparing carefully\" versus \"we are about to run out of fuel\" — was filled by deference to the four stripes rather than by arithmetic.\n\n**S4 — Check for dysfunction.** Textbook pattern: dissent expressed as mitigated speech (soft advisories, indirect mentions of fuel figures), no escalation when the advisories were not acted on, and a captain absorbed in one problem while the crew's concern never registered as a command-level input. The NTSB's probable-cause finding cited the captain's failure to monitor and respond to the fuel state, and — contributing — the other crew members' failure either to grasp the criticality or to successfully communicate their concern to the captain.\n\n**S5 — Cialdini defense (retrospective).** Q1 — genuine expert on this specific question? On flying the DC-8, yes; on the fuel-time tradeoff at that moment, the flight engineer's data outranked the captain's attention. Q2 — does the reasoning hold under independent scrutiny? No: independent arithmetic on fuel flow contradicted the plan to keep holding. Either N → discount the authority signal and demand the evidence. The evidence was on the panel.\n\n**S6 — Structural counter.** This is why United 173 matters more than most crashes: the NTSB explicitly recommended crew-coordination training addressing exactly this failure mode, and in 1981 United Airlines instituted the industry's first Crew Resource Management program, building on NASA research into flight-deck resource management. CRM operationalizes the skill's structural counters — juniors are trained and authorized to challenge assertively with escalating phraseology, and captains are trained to explicitly solicit input and treat it as data, not insubordination. Awareness alone was never expected to fix the bias; the industry installed structure. CRM subsequently spread to airlines worldwide and into medicine, firefighting, and maritime operations.\n\nThe mapped steps:\n1. Identify: captain as near-absolute authority; question = keep holding vs. land now; steep gradient suppressing challenge\n2. Expertise vs. positional: captain expert in flying, but the binding constraint (fuel) was the flight engineer's instrumented domain\n3. Examine reasoning: cautious troubleshooting was sound only under an unstated abundant-time assumption; the gap was filled by rank\n4. Dysfunction check: mitigated speech, hints not acted on, no escalation — dissent structurally muted\n5. Cialdini defense: Q2 fails — independent fuel arithmetic contradicted the plan; authority signal should have been discounted\n6. Structural counter: NTSB recommendation → United's 1981 CRM program → industry-wide assertive-challenge and input-solicitation protocols\n\nPrimary source: National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The CRM aftermath is documented in Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.), *Crew Resource Management.* Academic Press.\n\nFile v1.0.4:skill-card.md\n\n## Description: <br>\nDetects authority-bias dynamics and helps agents evaluate whether deference is justified by specific domain expertise and independently sound reasoning. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and decision-support agents use this skill to examine high-credential or hierarchical recommendations in boards, investment committees, medical or technical discussions, negotiations, and organizations where senior voices may suppress dissent. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Users may apply the guidance to sensitive business, medical, or organizational decisions. <br>\nMitigation: Keep private details out of manually added examples and verify cited claims independently for high-stakes use. <br>\nRisk: The skill may over-discount legitimate expertise if the user omits context about the authority's domain-specific track record. <br>\nMitigation: Use the skill's domain-expertise and independent-reasoning checks before reducing the weight of an expert recommendation. <br>\n\n\n## Reference(s): <br>\n- [Authority Bias skill page](https://clawhub.ai/deciqai/skills/authority-bias) <br>\n- [Primary sources for authority bias](references/sources.md) <br>\n- [Milgram 1963, Cialdini, Theranos, and FTX example](examples/milgram-1963-cialdini-systematization-theranos-ftx.md) <br>\n- [United 173 cockpit authority-gradient example](examples/united-173-1978-cockpit-authority-gradient-and-crm.md) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [Text, Markdown, Analysis, Guidance] <br>\n**Output Format:** [Markdown text with a structured authority-bias analysis template] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May ask stepwise questions and stop at explicit wait points when coaching novices.] <br>\n\n## Skill Version(s): <br>\n1.0.4 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.3: 6 files, 11834 bytes\n\nFiles: examples/milgram-1963-cialdini-systematization-theranos-ftx.md (5467b), examples/united-173-1978-cockpit-authority-gradient-and-crm.md (4990b), references/sources.md (1711b), skill-card.md (2415b), SKILL.md (7855b), _meta.json (133b)\n\nFile v1.0.3:SKILL.md\n\n---\nname: authority-bias\ndescription: \"Activate when: someone says 'should we trust the expert on this?', 'they're a professor/CEO/MD so they must be right', 'I don't want to push back on them', 'we deferred to the board and it blew up', or is evaluating a charismatic founder, assessing due diligence on a high-credential deal, or building a team where senior voices dominate.\nDo NOT activate when: the authority is a verified domain expert on the exact question at hand and the decision is time-critical; or when the user is asking a general persuasion question with no authority dynamic involved.\"\n---\n\n# Authority Bias\n\n## Overview\n\n**Authority bias** is the automatic tendency to comply with directives from authority figures — often overriding one's own judgment, ethics, or evidence. It operates beneath conscious deliberation: people substitute the authority's judgment for their own rather than weighing it.\n\nMilgram (1963) showed 65% of ordinary adults administered apparent 450V shocks when instructed by a lab-coated experimenter — against expert predictions of 1-2%. Cialdini (1984) systematized three authority triggers: titles, attire, and trappings — all of which fire regardless of actual expertise.\n\nKey distinction: **positional authority** (CEO, board chair, VC partner) vs. **domain expertise** (demonstrated track record on this specific question type). The bias treats them as identical. The skill does not.\n\nComposes with `social-proof`, `reciprocity`, `critical-thinking`, `signaling-games`, `dunning-kruger`.\n\n## When to Use\n\n- Boards, investment committees, due diligence on high-credential founders\n- Medical or technical decisions where expert opinion dominates discussion\n- Org decisions where senior voices suppress dissent; 360-feedback design\n- Negotiations where counterparty leverages credentials or position\n\n**Not when:** authority is verified domain expert on the exact question and speed matters.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** concrete case → run The Process directly.\n- **Coach mode:** unfamiliar or no case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: before deferring, ask whether they're a genuine expert on this specific question and whether their reasoning holds up independently.\n2. Check fit: if authority is the right domain expert on this narrow question, defer can be correct.\n3. Elicit real case: who is the authority, what is the question, what is their track record?\n> **[WAIT — do not advance until user responds]**\n4. Run S1-S5 one step at a time with their input.\n> **[WAIT — do not advance until user responds]**\n5. Close: name the insight — calibrated weighting + structural counters if relevant.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**S1 — Identify:** Authority (person/institution) · question/decision · authority signals (title, attire, credentials) · pressure to defer.\n\n**S2 — Expertise vs. positional:** Actual track record on this question type? Positional (CEO, chair) or domain expertise (years operating here)? Would a peer expert agree?\n\n**S3 — Examine reasoning:** Is the reasoning sound independent of who's stating it? Are logical gaps being filled by authority signals?\n\n**S4 — Check for dysfunction:** Is dissent suppressed? Are decisions justified by \"X said so\"? Is positional authority conflated with domain expertise?\n\n**S5 — Cialdini defense:** (Q1) Genuine expert on this specific question? (Q2) Reasoning holds under independent scrutiny? Both Y → appropriate weight. Either N → discount signal, demand evidence.\n\n**S6 — Structural counter (if you are the authority):** Explicitly invite dissent (\"give me 3 reasons against\") · separate suggestions from directives · install red team / premortem.\n\n## Output: Authority-Bias Analysis\n```\nAuthority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:\n```\n\n*→ Method in Action: [Milgram 1963 + Cialdini Systematization + Theranos/FTX](examples/milgram-1963-cialdini-systematization-theranos-ftx.md) · [United 173 and the Cockpit Authority Gradient](examples/united-173-1978-cockpit-authority-gradient-and-crm.md)*\n\n## Pack: Authority-Bias Application Patterns\n\n| Situation | Defensive question |\n|---|---|\n| Investment in startup (CEO / VC brand) | Have they demonstrated this specific business model works? |\n| Board member's opinion | Is this question actually in their domain expertise? |\n| Medical recommendation | What's the evidence base, independent of the white coat? |\n| Senior engineer's design decision | Has this approach worked in our specific context before? |\n| Founder's product vision | Are users requesting this, or is it founder preference? |\n\n## Applying It Well\n\n- Awareness doesn't prevent deference — structural countermeasures are required.\n- When you are the authority, your words are amplified beyond intent; operate accordingly.\n- Cialdini defense: (a) genuine expert on this question? (b) reasoning holds independently?\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"They're smart; we should listen\" | Smart in general doesn't mean smart on this specific question. Apply the domain test. |\n| [D] \"Everyone respects them\" | Cialdini's social proof reinforcing authority. Both biases compound. Separate them. |\n| [D] \"They have a great track record\" | In what specifically? Apply the question scope check. |\n| [D] \"It would be rude to push back\" | Compliance despite better judgment is the bias operating. Rudeness is not the actual cost. |\n| [D] \"I'm not in a position to disagree\" | If you have contradicting evidence or reasoning, you are. The framing is authority bias self-reinforcing. |\n| [D] \"If they're wrong, it'll show in due diligence\" | Theranos and FTX survived layers of due diligence by leveraging authority to suppress scrutiny. |\n| [D] \"I'm too junior\" | Junior people often have unfiltered observations. Suppressing them is where structural failures occur. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\n- Decision made primarily because an authority said so, not because the reasoning holds\n- Dissent suppressed — junior people silent, mid-managers deferring without pushback\n- Due diligence shortened due to authority signals; credentials conflated with domain expertise\n- No red team / premortem / devil's advocate on a high-stakes call\n\n## Verification\n\n- [ ] Authority's track record on this specific question examined\n- [ ] Positional authority distinguished from domain expertise\n- [ ] Reasoning independently examined; at least one dissenting expert sought\n- [ ] Cialdini two-question defense applied\n- [ ] If org: psychological safety mechanisms in place\n- [ ] If you are the authority: structural dissent-encouragement operational\n- [ ] Decision documented with reasoning, not just authority attribution\n\n*Part of **deciqAI Knowledge Skills** — 163 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/skills/authority-bias?utm_source=clawhub&utm_medium=marketplace&utm_campaign=knowledge-skills&utm_content=authority-bias** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.3:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"authority-bias\",\n  \"version\": \"1.0.3\",\n  \"publishedAt\": 1783482639152\n}\n\nFile v1.0.3:references/sources.md\n\n# Sources — authority-bias\n\n> *Primary sources for the [authority-bias](../SKILL.md) skill.*\n\n- Milgram, S. (1963). \"Behavioral Study of Obedience.\" *Journal of Abnormal and Social Psychology*, 67(4), 371-378. The foundational experiment.\n- Milgram, S. (1974). *Obedience to Authority: An Experimental View.* Harper & Row. ISBN 978-0061765216. The comprehensive treatment.\n- Cialdini, R. B. (1984/2006). *Influence: The Psychology of Persuasion.* HarperBusiness. ISBN 978-0061241895. Ch. 6.\n- Burger, J. M. (2009). \"Replicating Milgram: Would People Still Obey Today?\" *American Psychologist*, 64(1), 1-11. Modern replication.\n- Edmondson, A. C. (1999). \"Psychological Safety and Learning Behavior in Work Teams.\" *Administrative Science Quarterly*, 44(2), 350-383.\n- Janis, I. L. (1972). *Victims of Groupthink.* Houghton Mifflin. ISBN 978-0395140024. Group dysfunction and authority dynamics.\n- Carreyrou, J. (2018). *Bad Blood: Secrets and Lies in a Silicon Valley Startup.* Knopf. ISBN 978-1524731656. The Theranos case.\n- Lewis, M. (2023). *Going Infinite: The Rise and Fall of a New Tycoon.* W. W. Norton. ISBN 978-1324074335. The FTX case.\n- Heath, C. & Heath, D. (2013). *Decisive: How to Make Better Choices in Life and Work.* Crown. Includes authority-bias countermeasures.\n- National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The cockpit authority gradient case that led to Crew Resource Management.\n- Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.) (2010). *Crew Resource Management.* Academic Press. CRM as a structural counter to cockpit authority bias.\n\nFile v1.0.3:examples/milgram-1963-cialdini-systematization-theranos-ftx.md\n\n# Method in Action: Milgram 1963 + Cialdini Systematization + Theranos/FTX\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nThe **Milgram 1963 experiment** is the foundational empirical case. The contrast between psychologist prediction (1-2% compliance) and actual result (65%) is the canonical demonstration of how little of human behavior is governed by the rational, ethical, individual judgment that we believe we possess and that observers believe we possess. Under authority pressure, most ordinary adults will administer apparent severe harm — not because they are bad people, but because authority compliance is a deeply embedded social instinct.\n\nMilgram's 1974 follow-up book provided the systematic treatment, including the 18+ variations that mapped the conditions of authority compliance. The pattern: legitimacy of authority, presence of authority, absence of dissenting peers, distance from the victim all increase compliance. The reverse conditions decrease it. The framework has been validated in cross-cultural replications, including Burger's 2009 partial replication.\n\n**Robert Cialdini's 1984 systematization** placed authority among the six universal weapons of influence. Cialdini's contribution was operational: identifying the specific signals (titles, attire, trappings) that trigger compliance and the defensive question structure (is this authority a real expert here?) that resists it.\n\n**Modern catastrophic examples** include:\n\n**Theranos (founded 2003, collapsed 2018).** Elizabeth Holmes recruited a board with extraordinary authority signals — former Secretaries of State (Kissinger, Shultz), Stanford-affiliated medical figures, decorated military officers. Investors deferred. Walgreens, Safeway, the Department of Defense made commitments based on authority signals rather than scientific due diligence. The actual technology never worked. Patients received inaccurate medical results. Total losses: billions of dollars; loss of trust in lab innovation; criminal convictions for Holmes. The authority bias suppressed due diligence that would have caught the fraud.\n\n**FTX (founded 2019, collapsed November 2022).** Sam Bankman-Fried carried MIT credentials, prominent VC endorsements (Sequoia, Tiger Global, Paradigm), and political connections (largest political donor in many cycles). Investors and customers deferred. The collapse revealed $8 billion+ of customer funds had been misappropriated; the company had no functional risk management or auditing; the leadership team was making decisions in group chats with no oversight. The authority signals had suppressed the scrutiny that would have caught the fraud earlier. Many of the \"smart money\" investors (including major hedge funds and family offices) lost their entire investments.\n\nThe pattern: organizations that should have caught these frauds did not, because authority signals were strong and the structural countermeasures (independent due diligence, demanding evidence rather than credentials, applying skepticism to charismatic founders) were absent. Authority bias isn't a peripheral cognitive quirk; it's a load-bearing failure mode in modern finance and business.\n\nThe framework has been applied operationally in:\n\n**Medical decision-making.** Modern medicine has moved away from the \"white coat says so\" model toward shared decision-making, evidence-based protocols, and second-opinion practices. The structural changes are explicit authority-bias countermeasures.\n\n**Aviation crew resource management.** After several crashes attributed to captain authority overriding junior officer concerns (e.g., the 1977 Tenerife disaster), commercial aviation adopted Crew Resource Management protocols. These explicitly authorize junior officers to challenge captain decisions and require captains to actively solicit input.\n\n**Investment committees.** Sophisticated investment processes now include explicit \"kill the deal\" sessions where someone is assigned to argue against the investment, regardless of the partners' enthusiasm.\n\n**Organizational change.** Psychological safety research (Edmondson 1999) has demonstrated that teams where dissent is welcome substantially outperform teams where authority bias is unchecked. Google's Project Aristotle (2012-2015) found psychological safety to be the strongest predictor of team performance.\n\nThree operational lessons:\n\n**First, authority bias is automatic and survives explicit awareness.** Knowing about Milgram does not prevent you from deferring to authority. The cognitive mechanism is too deeply embedded. Structural countermeasures (premortems, devil's advocates, red teams, psychological safety) are required.\n\n**Second, distinguishing expertise from positional authority is the discipline.** A renowned investor's opinion on operations is positional, not expert. A charismatic CEO's claim about their technology is positional, not domain-expert if the CEO is not a technologist. The Cialdini two-question defense is the operational test.\n\n**Third, when you are the authority, you bear specific responsibility.** Your words and presence are systematically amplified, often beyond your intent. The structural counters (explicit dissent encouragement, separating suggestion from directive, asking explicitly for opposing views) compensate for the asymmetric weight. Operating as if your words have normal weight is operating with unrecognized power, and that power produces predictable errors.\n\nFile v1.0.3:examples/united-173-1978-cockpit-authority-gradient-and-crm.md\n\n# Method in Action: United 173 and the Cockpit Authority Gradient (1978 → 1981)\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nOn December 28, 1978, United Airlines Flight 173, a DC-8 inbound to Portland, Oregon, experienced a landing gear anomaly on approach — an abnormal indication and an unusual noise when the gear extended. The captain entered a holding pattern to troubleshoot and to prepare the cabin for a possible emergency landing. The airplane held for roughly an hour. During that hour, the fuel ran out. The DC-8 crash-landed in a wooded suburban area about six miles short of the airport. Ten people aboard died, including the flight engineer. Post-crash examination showed the landing gear had been down and functional the whole time.\n\n**S1 — Identify.** The authority: a highly experienced captain, backed by the era's cockpit culture in which the captain's judgment was close to unchallengeable. The question: when to stop troubleshooting and land. The pressure to defer: a steep authority gradient — the first officer and flight engineer were structurally subordinate, and the norm was to advise, never to countermand.\n\n**S2 — Expertise vs. positional.** The captain was a genuine domain expert as an aviator. But the specific question in play — remaining fuel versus remaining time — was squarely instrumented on the flight engineer's panel. The flight engineer had the better read on the binding constraint. Positional authority outweighed the crew member with the most relevant information.\n\n**S3 — Examine reasoning.** The captain's reasoning was locally sound: an unresolved gear indication justifies caution and cabin preparation. But it silently assumed time was abundant. The NTSB found that both junior crew members made advisory remarks about the deteriorating fuel state, yet neither converted those hints into an assertive, unambiguous challenge. The logical gap — \"we are preparing carefully\" versus \"we are about to run out of fuel\" — was filled by deference to the four stripes rather than by arithmetic.\n\n**S4 — Check for dysfunction.** Textbook pattern: dissent expressed as mitigated speech (soft advisories, indirect mentions of fuel figures), no escalation when the advisories were not acted on, and a captain absorbed in one problem while the crew's concern never registered as a command-level input. The NTSB's probable-cause finding cited the captain's failure to monitor and respond to the fuel state, and — contributing — the other crew members' failure either to grasp the criticality or to successfully communicate their concern to the captain.\n\n**S5 — Cialdini defense (retrospective).** Q1 — genuine expert on this specific question? On flying the DC-8, yes; on the fuel-time tradeoff at that moment, the flight engineer's data outranked the captain's attention. Q2 — does the reasoning hold under independent scrutiny? No: independent arithmetic on fuel flow contradicted the plan to keep holding. Either N → discount the authority signal and demand the evidence. The evidence was on the panel.\n\n**S6 — Structural counter.** This is why United 173 matters more than most crashes: the NTSB explicitly recommended crew-coordination training addressing exactly this failure mode, and in 1981 United Airlines instituted the industry's first Crew Resource Management program, building on NASA research into flight-deck resource management. CRM operationalizes the skill's structural counters — juniors are trained and authorized to challenge assertively with escalating phraseology, and captains are trained to explicitly solicit input and treat it as data, not insubordination. Awareness alone was never expected to fix the bias; the industry installed structure. CRM subsequently spread to airlines worldwide and into medicine, firefighting, and maritime operations.\n\nThe mapped steps:\n1. Identify: captain as near-absolute authority; question = keep holding vs. land now; steep gradient suppressing challenge\n2. Expertise vs. positional: captain expert in flying, but the binding constraint (fuel) was the flight engineer's instrumented domain\n3. Examine reasoning: cautious troubleshooting was sound only under an unstated abundant-time assumption; the gap was filled by rank\n4. Dysfunction check: mitigated speech, hints not acted on, no escalation — dissent structurally muted\n5. Cialdini defense: Q2 fails — independent fuel arithmetic contradicted the plan; authority signal should have been discounted\n6. Structural counter: NTSB recommendation → United's 1981 CRM program → industry-wide assertive-challenge and input-solicitation protocols\n\nPrimary source: National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The CRM aftermath is documented in Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.), *Crew Resource Management.* Academic Press.\n\nFile v1.0.3:skill-card.md\n\n## Description: <br>\nDetect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, developers, and decision teams use this skill to test whether authority signals are being over-weighted in boards, investment committees, medical or technical decisions, organizations, negotiations, and due diligence. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Users could treat the skill as an authority for security, medical, or other high-stakes decisions. <br>\nMitigation: Use it as a reasoning aid and seek independent evidence or qualified review before acting on high-stakes conclusions. <br>\nRisk: The skill may encourage challenging authority claims in situations where context-sensitive judgment is needed. <br>\nMitigation: Review the source links and examples to assess the underlying psychology claims and apply the skill only when authority dynamics are relevant. <br>\n\n\n## Reference(s): <br>\n- [Sources - authority-bias](references/sources.md) <br>\n- [Milgram 1963 + Cialdini Systematization + Theranos/FTX](examples/milgram-1963-cialdini-systematization-theranos-ftx.md) <br>\n- [United 173 and the Cockpit Authority Gradient](examples/united-173-1978-cockpit-authority-gradient-and-crm.md) <br>\n- [deciqAI Knowledge Skills repository](https://github.com/deciqAI/knowledge-skills) <br>\n- [Authority Bias skill page](https://clawhub.ai/deciqai/skills/authority-bias) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, guidance] <br>\n**Output Format:** [Markdown analysis template with structured decision prompts] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [May pause for user input during coach-mode steps before completing the analysis.] <br>\n\n## Skill Version(s): <br>\n1.0.3 (source: server release evidence) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.2: 5 files, 8872 bytes\n\nFiles: examples/milgram-1963-cialdini-systematization-theranos-ftx.md (5467b), references/sources.md (1286b), skill-card.md (2254b), SKILL.md (7740b), _meta.json (133b)\n\nFile v1.0.2:SKILL.md\n\n---\nname: authority-bias\ndescription: \"Activate when: someone says 'should we trust the expert on this?', 'they're a professor/CEO/MD so they must be right', 'I don't want to push back on them', 'we deferred to the board and it blew up', or is evaluating a charismatic founder, assessing due diligence on a high-credential deal, or building a team where senior voices dominate.\nDo NOT activate when: the authority is a verified domain expert on the exact question at hand and the decision is time-critical; or when the user is asking a general persuasion question with no authority dynamic involved.\"\n---\n\n# Authority Bias\n\n## Overview\n\n**Authority bias** is the automatic tendency to comply with directives from authority figures — often overriding one's own judgment, ethics, or evidence. It operates beneath conscious deliberation: people substitute the authority's judgment for their own rather than weighing it.\n\nMilgram (1963) showed 65% of ordinary adults administered apparent 450V shocks when instructed by a lab-coated experimenter — against expert predictions of 1-2%. Cialdini (1984) systematized three authority triggers: titles, attire, and trappings — all of which fire regardless of actual expertise.\n\nKey distinction: **positional authority** (CEO, board chair, VC partner) vs. **domain expertise** (demonstrated track record on this specific question type). The bias treats them as identical. The skill does not.\n\nComposes with `social-proof`, `reciprocity`, `critical-thinking`, `signaling-games`, `dunning-kruger`.\n\n## When to Use\n\n- Boards, investment committees, due diligence on high-credential founders\n- Medical or technical decisions where expert opinion dominates discussion\n- Org decisions where senior voices suppress dissent; 360-feedback design\n- Negotiations where counterparty leverages credentials or position\n\n**Not when:** authority is verified domain expert on the exact question and speed matters.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** concrete case → run The Process directly.\n- **Coach mode:** unfamiliar or no case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: before deferring, ask whether they're a genuine expert on this specific question and whether their reasoning holds up independently.\n2. Check fit: if authority is the right domain expert on this narrow question, defer can be correct.\n3. Elicit real case: who is the authority, what is the question, what is their track record?\n> **[WAIT — do not advance until user responds]**\n4. Run S1-S5 one step at a time with their input.\n> **[WAIT — do not advance until user responds]**\n5. Close: name the insight — calibrated weighting + structural counters if relevant.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**S1 — Identify:** Authority (person/institution) · question/decision · authority signals (title, attire, credentials) · pressure to defer.\n\n**S2 — Expertise vs. positional:** Actual track record on this question type? Positional (CEO, chair) or domain expertise (years operating here)? Would a peer expert agree?\n\n**S3 — Examine reasoning:** Is the reasoning sound independent of who's stating it? Are logical gaps being filled by authority signals?\n\n**S4 — Check for dysfunction:** Is dissent suppressed? Are decisions justified by \"X said so\"? Is positional authority conflated with domain expertise?\n\n**S5 — Cialdini defense:** (Q1) Genuine expert on this specific question? (Q2) Reasoning holds under independent scrutiny? Both Y → appropriate weight. Either N → discount signal, demand evidence.\n\n**S6 — Structural counter (if you are the authority):** Explicitly invite dissent (\"give me 3 reasons against\") · separate suggestions from directives · install red team / premortem.\n\n## Output: Authority-Bias Analysis\n```\nAuthority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:\n```\n\n*→ Method in Action: [Milgram 1963 + Cialdini Systematization + Theranos/FTX](examples/milgram-1963-cialdini-systematization-theranos-ftx.md)*\n\n## Pack: Authority-Bias Application Patterns\n\n| Situation | Defensive question |\n|---|---|\n| Investment in startup (CEO / VC brand) | Have they demonstrated this specific business model works? |\n| Board member's opinion | Is this question actually in their domain expertise? |\n| Medical recommendation | What's the evidence base, independent of the white coat? |\n| Senior engineer's design decision | Has this approach worked in our specific context before? |\n| Founder's product vision | Are users requesting this, or is it founder preference? |\n\n## Applying It Well\n\n- Awareness doesn't prevent deference — structural countermeasures are required.\n- When you are the authority, your words are amplified beyond intent; operate accordingly.\n- Cialdini defense: (a) genuine expert on this question? (b) reasoning holds independently?\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"They're smart; we should listen\" | Smart in general doesn't mean smart on this specific question. Apply the domain test. |\n| [D] \"Everyone respects them\" | Cialdini's social proof reinforcing authority. Both biases compound. Separate them. |\n| [D] \"They have a great track record\" | In what specifically? Apply the question scope check. |\n| [D] \"It would be rude to push back\" | Compliance despite better judgment is the bias operating. Rudeness is not the actual cost. |\n| [D] \"I'm not in a position to disagree\" | If you have contradicting evidence or reasoning, you are. The framing is authority bias self-reinforcing. |\n| [D] \"If they're wrong, it'll show in due diligence\" | Theranos and FTX survived layers of due diligence by leveraging authority to suppress scrutiny. |\n| [D] \"I'm too junior\" | Junior people often have unfiltered observations. Suppressing them is where structural failures occur. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\n- Decision made primarily because an authority said so, not because the reasoning holds\n- Dissent suppressed — junior people silent, mid-managers deferring without pushback\n- Due diligence shortened due to authority signals; credentials conflated with domain expertise\n- No red team / premortem / devil's advocate on a high-stakes call\n\n## Verification\n\n- [ ] Authority's track record on this specific question examined\n- [ ] Positional authority distinguished from domain expertise\n- [ ] Reasoning independently examined; at least one dissenting expert sought\n- [ ] Cialdini two-question defense applied\n- [ ] If org: psychological safety mechanisms in place\n- [ ] If you are the authority: structural dissent-encouragement operational\n- [ ] Decision documented with reasoning, not just authority attribution\n\n*Part of **deciqAI Knowledge Skills** — 163 open-source thinking skills that make rigor executable for AI agents. The same skills power every deciqAI agent, which runs them autonomously to operate your company. **See it run → https://www.deciqai.com/skills/authority-bias?utm_source=clawhub&utm_medium=marketplace&utm_campaign=knowledge-skills&utm_content=authority-bias** · ⭐ Star the repo → https://github.com/deciqAI/knowledge-skills · Contributions welcome.*\n\nFile v1.0.2:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"authority-bias\",\n  \"version\": \"1.0.2\",\n  \"publishedAt\": 1783471097507\n}\n\nFile v1.0.2:references/sources.md\n\n# Sources — authority-bias\n\n> *Primary sources for the [authority-bias](../SKILL.md) skill.*\n\n- Milgram, S. (1963). \"Behavioral Study of Obedience.\" *Journal of Abnormal and Social Psychology*, 67(4), 371-378. The foundational experiment.\n- Milgram, S. (1974). *Obedience to Authority: An Experimental View.* Harper & Row. ISBN 978-0061765216. The comprehensive treatment.\n- Cialdini, R. B. (1984/2006). *Influence: The Psychology of Persuasion.* HarperBusiness. ISBN 978-0061241895. Ch. 6.\n- Burger, J. M. (2009). \"Replicating Milgram: Would People Still Obey Today?\" *American Psychologist*, 64(1), 1-11. Modern replication.\n- Edmondson, A. C. (1999). \"Psychological Safety and Learning Behavior in Work Teams.\" *Administrative Science Quarterly*, 44(2), 350-383.\n- Janis, I. L. (1972). *Victims of Groupthink.* Houghton Mifflin. ISBN 978-0395140024. Group dysfunction and authority dynamics.\n- Carreyrou, J. (2018). *Bad Blood: Secrets and Lies in a Silicon Valley Startup.* Knopf. ISBN 978-1524731656. The Theranos case.\n- Lewis, M. (2023). *Going Infinite: The Rise and Fall of a New Tycoon.* W. W. Norton. ISBN 978-1324074335. The FTX case.\n- Heath, C. & Heath, D. (2013). *Decisive: How to Make Better Choices in Life and Work.* Crown. Includes authority-bias countermeasures.\n\nFile v1.0.2:examples/milgram-1963-cialdini-systematization-theranos-ftx.md\n\n# Method in Action: Milgram 1963 + Cialdini Systematization + Theranos/FTX\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nThe **Milgram 1963 experiment** is the foundational empirical case. The contrast between psychologist prediction (1-2% compliance) and actual result (65%) is the canonical demonstration of how little of human behavior is governed by the rational, ethical, individual judgment that we believe we possess and that observers believe we possess. Under authority pressure, most ordinary adults will administer apparent severe harm — not because they are bad people, but because authority compliance is a deeply embedded social instinct.\n\nMilgram's 1974 follow-up book provided the systematic treatment, including the 18+ variations that mapped the conditions of authority compliance. The pattern: legitimacy of authority, presence of authority, absence of dissenting peers, distance from the victim all increase compliance. The reverse conditions decrease it. The framework has been validated in cross-cultural replications, including Burger's 2009 partial replication.\n\n**Robert Cialdini's 1984 systematization** placed authority among the six universal weapons of influence. Cialdini's contribution was operational: identifying the specific signals (titles, attire, trappings) that trigger compliance and the defensive question structure (is this authority a real expert here?) that resists it.\n\n**Modern catastrophic examples** include:\n\n**Theranos (founded 2003, collapsed 2018).** Elizabeth Holmes recruited a board with extraordinary authority signals — former Secretaries of State (Kissinger, Shultz), Stanford-affiliated medical figures, decorated military officers. Investors deferred. Walgreens, Safeway, the Department of Defense made commitments based on authority signals rather than scientific due diligence. The actual technology never worked. Patients received inaccurate medical results. Total losses: billions of dollars; loss of trust in lab innovation; criminal convictions for Holmes. The authority bias suppressed due diligence that would have caught the fraud.\n\n**FTX (founded 2019, collapsed November 2022).** Sam Bankman-Fried carried MIT credentials, prominent VC endorsements (Sequoia, Tiger Global, Paradigm), and political connections (largest political donor in many cycles). Investors and customers deferred. The collapse revealed $8 billion+ of customer funds had been misappropriated; the company had no functional risk management or auditing; the leadership team was making decisions in group chats with no oversight. The authority signals had suppressed the scrutiny that would have caught the fraud earlier. Many of the \"smart money\" investors (including major hedge funds and family offices) lost their entire investments.\n\nThe pattern: organizations that should have caught these frauds did not, because authority signals were strong and the structural countermeasures (independent due diligence, demanding evidence rather than credentials, applying skepticism to charismatic founders) were absent. Authority bias isn't a peripheral cognitive quirk; it's a load-bearing failure mode in modern finance and business.\n\nThe framework has been applied operationally in:\n\n**Medical decision-making.** Modern medicine has moved away from the \"white coat says so\" model toward shared decision-making, evidence-based protocols, and second-opinion practices. The structural changes are explicit authority-bias countermeasures.\n\n**Aviation crew resource management.** After several crashes attributed to captain authority overriding junior officer concerns (e.g., the 1977 Tenerife disaster), commercial aviation adopted Crew Resource Management protocols. These explicitly authorize junior officers to challenge captain decisions and require captains to actively solicit input.\n\n**Investment committees.** Sophisticated investment processes now include explicit \"kill the deal\" sessions where someone is assigned to argue against the investment, regardless of the partners' enthusiasm.\n\n**Organizational change.** Psychological safety research (Edmondson 1999) has demonstrated that teams where dissent is welcome substantially outperform teams where authority bias is unchecked. Google's Project Aristotle (2012-2015) found psychological safety to be the strongest predictor of team performance.\n\nThree operational lessons:\n\n**First, authority bias is automatic and survives explicit awareness.** Knowing about Milgram does not prevent you from deferring to authority. The cognitive mechanism is too deeply embedded. Structural countermeasures (premortems, devil's advocates, red teams, psychological safety) are required.\n\n**Second, distinguishing expertise from positional authority is the discipline.** A renowned investor's opinion on operations is positional, not expert. A charismatic CEO's claim about their technology is positional, not domain-expert if the CEO is not a technologist. The Cialdini two-question defense is the operational test.\n\n**Third, when you are the authority, you bear specific responsibility.** Your words and presence are systematically amplified, often beyond your intent. The structural counters (explicit dissent encouragement, separating suggestion from directive, asking explicitly for opposing views) compensate for the asymmetric weight. Operating as if your words have normal weight is operating with unrecognized power, and that power produces predictable errors.\n\nFile v1.0.2:skill-card.md\n\n## Description: <br>\nDetect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and decision-making teams use this skill to evaluate situations where credentials, rank, or institutional authority may be distorting a decision. It helps calibrate deference against domain expertise, independent reasoning, and structural dissent checks. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: The skill may be mistaken for professional medical, investment, legal, or organizational advice in high-stakes authority-driven decisions. <br>\nMitigation: Treat outputs as decision-support guidance and verify conclusions against qualified domain expertise, evidence, and applicable professional review. <br>\nRisk: Users may include confidential organizational, medical, or investment details while applying the analysis. <br>\nMitigation: Minimize or anonymize sensitive details and avoid retaining them in persistent notes unless retention is intentional and approved. <br>\n\n\n## Reference(s): <br>\n- [Authority Bias on ClawHub](https://clawhub.ai/deciqai/skills/authority-bias) <br>\n- [Primary sources](references/sources.md) <br>\n- [Method in Action: Milgram 1963 + Cialdini Systematization + Theranos/FTX](examples/milgram-1963-cialdini-systematization-theranos-ftx.md) <br>\n\n\n## Skill Output: <br>\n**Output Type(s):** [text, markdown, guidance] <br>\n**Output Format:** [Markdown analysis with structured fields] <br>\n**Output Parameters:** [1D] <br>\n**Other Properties Related to Output:** [Markdown-only guidance; no executable behavior.] <br>\n\n## Skill Version(s): <br>\n1.0.2 (source: server release metadata) <br>\n\n## Ethical Considerations: <br>\nUsers should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment. <br>\n\nArchive v1.0.1: 5 files, 8842 bytes\n\nFiles: examples/milgram-1963-cialdini-systematization-theranos-ftx.md (5467b), references/sources.md (1286b), skill-card.md (2401b), SKILL.md (7633b), _meta.json (133b)\n\nFile v1.0.1:SKILL.md\n\n---\nname: authority-bias\ndescription: \"Activate when: someone says 'should we trust the expert on this?', 'they're a professor/CEO/MD so they must be right', 'I don't want to push back on them', 'we deferred to the board and it blew up', or is evaluating a charismatic founder, assessing due diligence on a high-credential deal, or building a team where senior voices dominate.\nDo NOT activate when: the authority is a verified domain expert on the exact question at hand and the decision is time-critical; or when the user is asking a general persuasion question with no authority dynamic involved.\"\n---\n\n# Authority Bias\n\n## Overview\n\n**Authority bias** is the automatic tendency to comply with directives from authority figures — often overriding one's own judgment, ethics, or evidence. It operates beneath conscious deliberation: people substitute the authority's judgment for their own rather than weighing it.\n\nMilgram (1963) showed 65% of ordinary adults administered apparent 450V shocks when instructed by a lab-coated experimenter — against expert predictions of 1-2%. Cialdini (1984) systematized three authority triggers: titles, attire, and trappings — all of which fire regardless of actual expertise.\n\nKey distinction: **positional authority** (CEO, board chair, VC partner) vs. **domain expertise** (demonstrated track record on this specific question type). The bias treats them as identical. The skill does not.\n\nComposes with [`social-proof`](../social-proof/SKILL.md), [`reciprocity`](../reciprocity/SKILL.md), [`critical-thinking`](../critical-thinking/SKILL.md), [`signaling-games`](../signaling-games/SKILL.md), [`dunning-kruger`](../dunning-kruger/SKILL.md).\n\n## When to Use\n\n- Boards, investment committees, due diligence on high-credential founders\n- Medical or technical decisions where expert opinion dominates discussion\n- Org decisions where senior voices suppress dissent; 360-feedback design\n- Negotiations where counterparty leverages credentials or position\n\n**Not when:** authority is verified domain expert on the exact question and speed matters.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** concrete case → run The Process directly.\n- **Coach mode:** unfamiliar or no case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: before deferring, ask whether they're a genuine expert on this specific question and whether their reasoning holds up independently.\n2. Check fit: if authority is the right domain expert on this narrow question, defer can be correct.\n3. Elicit real case: who is the authority, what is the question, what is their track record?\n> **[WAIT — do not advance until user responds]**\n4. Run S1-S5 one step at a time with their input.\n> **[WAIT — do not advance until user responds]**\n5. Close: name the insight — calibrated weighting + structural counters if relevant.\n> **[WAIT — do not advance until user responds]**\n\n## The Process\n\n**S1 — Identify:** Authority (person/institution) · question/decision · authority signals (title, attire, credentials) · pressure to defer.\n\n**S2 — Expertise vs. positional:** Actual track record on this question type? Positional (CEO, chair) or domain expertise (years operating here)? Would a peer expert agree?\n\n**S3 — Examine reasoning:** Is the reasoning sound independent of who's stating it? Are logical gaps being filled by authority signals?\n\n**S4 — Check for dysfunction:** Is dissent suppressed? Are decisions justified by \"X said so\"? Is positional authority conflated with domain expertise?\n\n**S5 — Cialdini defense:** (Q1) Genuine expert on this specific question? (Q2) Reasoning holds under independent scrutiny? Both Y → appropriate weight. Either N → discount signal, demand evidence.\n\n**S6 — Structural counter (if you are the authority):** Explicitly invite dissent (\"give me 3 reasons against\") · separate suggestions from directives · install red team / premortem.\n\n## Output: Authority-Bias Analysis\n```\nAuthority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:\n```\n\n*→ Method in Action: [Milgram 1963 + Cialdini Systematization + Theranos/FTX](examples/milgram-1963-cialdini-systematization-theranos-ftx.md)*\n\n## Pack: Authority-Bias Application Patterns\n\n| Situation | Defensive question |\n|---|---|\n| Investment in startup (CEO / VC brand) | Have they demonstrated this specific business model works? |\n| Board member's opinion | Is this question actually in their domain expertise? |\n| Medical recommendation | What's the evidence base, independent of the white coat? |\n| Senior engineer's design decision | Has this approach worked in our specific context before? |\n| Founder's product vision | Are users requesting this, or is it founder preference? |\n\n## Applying It Well\n\n- Awareness doesn't prevent deference — structural countermeasures are required.\n- When you are the authority, your words are amplified beyond intent; operate accordingly.\n- Cialdini defense: (a) genuine expert on this question? (b) reasoning holds independently?\n\n*→ Primary sources: [references/sources.md](references/sources.md)*\n\n## Common Rationalizations\n\n**[D] = designed upfront | [O] = observed in real use. [O] entries are more valuable.**\n\n| Fake move | Reality |\n|---|---|\n| [D] \"They're smart; we should listen\" | Smart in general doesn't mean smart on this specific question. Apply the domain test. |\n| [D] \"Everyone respects them\" | Cialdini's social proof reinforcing authority. Both biases compound. Separate them. |\n| [D] \"They have a great track record\" | In what specifically? Apply the question scope check. |\n| [D] \"It would be rude to push back\" | Compliance despite better judgment is the bias operating. Rudeness is not the actual cost. |\n| [D] \"I'm not in a position to disagree\" | If you have contradicting evidence or reasoning, you are. The framing is authority bias self-reinforcing. |\n| [D] \"If they're wrong, it'll show in due diligence\" | Theranos and FTX survived layers of due diligence by leveraging authority to suppress scrutiny. |\n| [D] \"I'm too junior\" | Junior people often have unfiltered observations. Suppressing them is where structural failures occur. |\n| *→ Add [O] entries here after each real use — paste the actual failure pattern* | *What went wrong and why* |\n\n## Red Flags\n\n- Decision made primarily because an authority said so, not because the reasoning holds\n- Dissent suppressed — junior people silent, mid-managers deferring without pushback\n- Due diligence shortened due to authority signals; credentials conflated with domain expertise\n- No red team / premortem / devil's advocate on a high-stakes call\n\n## Verification\n\n- [ ] Authority's track record on this specific question examined\n- [ ] Positional authority distinguished from domain expertise\n- [ ] Reasoning independently examined; at least one dissenting expert sought\n- [ ] Cialdini two-question defense applied\n- [ ] If org: psychological safety mechanisms in place\n- [ ] If you are the authority: structural dissent-encouragement operational\n- [ ] Decision documented with reasoning, not just authority attribution\n\n*Part of **deciqAI Knowledge Skills** — open-source thinking skills that make rigor executable for AI agents. Built by deciqAI · https://deciqai.com · Contributions welcome — see the template at the repo root.*\n\nFile v1.0.1:_meta.json\n\n{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"authority-bias\",\n  \"version\": \"1.0.1\",\n  \"publishedAt\": 1783456206303\n}\n\nFile v1.0.1:references/sources.md\n\n# Sources — authority-bias\n\n> *Primary sources for the [authority-bias](../SKILL.md) skill.*\n\n- Milgram, S. (1963). \"Behavioral Study of Obedience.\" *Journal of Abnormal and Social Psychology*, 67(4), 371-378. The foundational experiment.\n- Milgram, S. (1974). *Obedience to Authority: An Experimental View.* Harper & Row. ISBN 978-0061765216. The comprehensive treatment.\n- Cialdini, R. B. (1984/2006). *Influence: The Psychology of Persuasion.* HarperBusiness. ISBN 978-0061241895. Ch. 6.\n- Burger, J. M. (2009). \"Replicating Milgram: Would People Still Obey Today?\" *American Psychologist*, 64(1), 1-11. Modern replication.\n- Edmondson, A. C. (1999). \"Psychological Safety and Learning Behavior in Work Teams.\" *Administrative Science Quarterly*, 44(2), 350-383.\n- Janis, I. L. (1972). *Victims of Groupthink.* Houghton Mifflin. ISBN 978-0395140024. Group dysfunction and authority dynamics.\n- Carreyrou, J. (2018). *Bad Blood: Secrets and Lies in a Silicon Valley Startup.* Knopf. ISBN 978-1524731656. The Theranos case.\n- Lewis, M. (2023). *Going Infinite: The Rise and Fall of a New Tycoon.* W. W. Norton. ISBN 978-1324074335. The FTX case.\n- Heath, C. & Heath, D. (2013). *Decisive: How to Make Better Choices in Life and Work.* Crown. Includes authority-bias countermeasures.\n\nFile v1.0.1:examples/milgram-1963-cialdini-systematization-theranos-ftx.md\n\n# Method in Action: Milgram 1963 + Cialdini Systematization + Theranos/FTX\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nThe **Milgram 1963 experiment** is the foundational empirical case. The contrast between psychologist prediction (1-2% compliance) and actual result (65%) is the canonical demonstration of how little of human behavior is governed by the rational, ethical, individual judgment that we believe we possess and that observers believe we possess. Under authority pressure, most ordinary adults will administer apparent severe harm — not because they are bad people, but because authority compliance is a deeply embedded social instinct.\n\nMilgram's 1974 follow-up book provided the systematic treatment, including the 18+ variations that mapped the conditions of authority compliance. The pattern: legitimacy of authority, presence of authority, absence of dissenting peers, distance from the victim all increase compliance. The reverse conditions decrease it. The framework has been validated in cross-cultural replications, including Burger's 2009 partial replication.\n\n**Robert Cialdini's 1984 systematization** placed authority among the six universal weapons of influence. Cialdini's contribution was operational: identifying the specific signals (titles, attire, trappings) that trigger compliance and the defensive question structure (is this authority a real expert here?) that resists it.\n\n**Modern catastrophic examples** include:\n\n**Theranos (founded 2003, collapsed 2018).** Elizabeth Holmes recruited a board with extraordinary authority signals — former Secretaries of State (Kissinger, Shultz), Stanford-affiliated medical figures, decorated military officers. Investors deferred. Walgreens, Safeway, the Department of Defense made commitments based on authority signals rather than scientific due diligence. The actual technology never worked. Patients received inaccurate medical results. Total losses: billions of dollars; loss of trust in lab innovation; criminal convictions for Holmes. The authority bias suppressed due diligence that would have caught the fraud.\n\n**FTX (founded 2019, collapsed November 2022).** Sam Bankman-Fried carried MIT credentials, prominent VC endorsements (Sequoia, Tiger Global, Paradigm), and political connections (largest political donor in many cycles). Investors and customers deferred. The collapse revealed $8 billion+ of customer funds had been misappropriated; the company had no functional risk management or auditing; the leadership team was making decisions in group chats with no oversight. The authority signals had suppressed the scrutiny that would have caught the fraud earlier. Many of the \"smart money\" investors (including major hedge funds and family offices) lost their entire investments.\n\nThe pattern: organizations that should have caught these frauds did not, because authority signals were strong and the structural countermeasures (independent due diligence, demanding evidence rather than credentials, applying skepticism to charismatic founders) were absent. Authority bias isn't a peripheral cognitive quirk; it's a load-bearing failure mode in modern finance and business.\n\nThe framework has been applied operationally in:\n\n**Medical decision-making.** Modern medicine has moved away from the \"white coat says so\" model toward shared decision-making, evidence-based protocols, and second-opinion practices. The structural changes are explicit authority-bias countermeasures.\n\n**Aviation crew resource management.** After several crashes attributed to captain authority overriding junior officer concerns (e.g., the 1977 Tenerife disaster), commercial aviation adopted Crew Resource Management protocols. These explicitly authorize junior officers to challenge captain decisions and require captains to actively solicit input.\n\n**Investment committees.** Sophisticated investment processes now include explicit \"kill the deal\" sessions where someone is assigned to argue against the investment, regardless of the partners' enthusiasm.\n\n**Organizational change.** Psychological safety research (Edmondson 1999) has demonstrated that teams where dissent is welcome substantially outperform teams where authority bias is unchecked. Google's Project Aristotle (2012-2015) found psychological safety to be the strongest predictor of team performance.\n\nThree operational lessons:\n\n**First, authority bias is automatic and survives explicit awareness.** Knowing about Milgram does not prevent you from deferring to authority. The cognitive mechanism is too deeply embedded. Structural countermeasures (premortems, devil's advocates, red teams, psychological safety) are required.\n\n**Second, distinguishing expertise from positional authority is the discipline.** A renowned investor's opinion on operations is positional, not expert. A charismatic CEO's claim about their technology is positional, not domain-expert if the CEO is not a technologist. The Cialdini two-question defense is the operational test.\n\n**Third, when you are the authority, you bear specific responsibility.** Your words and presence are systematically amplified, often beyond your intent. The structural counters (explicit dissent encouragement, separating suggestion from directive, asking explicitly for opposing views) compensate for the asymmetric weight. Operating as if your words have normal weight is operating with unrecognized power, and that power produces predictable errors.\n\nFile v1.0.1:skill-card.md\n\n## Description: <br>\nDetect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. <br>\n\nThis skill is ready for commercial/non-commercial use. <br>\n\n## Publisher: <br>\n[deciqai](https://clawhub.ai/user/deciqai) <br>\n\n### License/Terms of Use: <br>\nMIT-0 <br>\n\n\n## Use Case: <br>\nEmployees, external users, and decision-makers use this skill to examine whether authority signals are being over-weighted in board, investment, medical, technical, organizational, or negotiation decisions. It helps distinguish positional authority from domain expertise and prompts independent review of reasoning before deferring. <br>\n\n### Deployment Geography for Use: <br>\nGlobal <br>\n\n## Known Risks and Mitigations: <br>\nRisk: Users may treat advisory decision-making guidance as professional, medical, legal, or financial advice. <br>\nMitigation: Use the skill as a thinking aid and verify high-stakes decisions with qualified professionals, independent evidence, and documented reasoning. <br>\nRisk: The skill can be misapplied to discount a verified domain expert when the authority is genuinely qualified on the exact question and time is critical. <br>\nMitigation: Apply the skill's stated non-use condition: defer can be appropriate when the authority is a verified domain expert on the narrow question and speed matters. <br>\n\n\n## Reference(s): <br>\n- [Sources - authority-bias](references/sources.md) <br>\n- [M\n\nArchive v1.0.0: 5 files, 8833 bytes\n\nFiles: examples/milgram-1963-cialdini-systematization-theranos-ftx.md (5467b), references/sources.md (1286b), skill-card.md (2251b), SKILL.md (7633b), _meta.json (133b)","readmeExcerpt":"Skill: Authority Bias Owner: deciqai Summary: Detect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. Tags: latest:1.0.6 Version history: v1.0.6 | 2026-07-16T17:52:10.284Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/authority-bias.json) v1.0.5 | 2026-07-10T10:24:08.661Z | user Ad","codeSnippets":[],"executableExamples":[{"language":"text","snippet":"Authority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:"},{"language":"text","snippet":"Authority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:"},{"language":"text","snippet":"Authority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:"},{"language":"text","snippet":"Authority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:"},{"language":"text","snippet":"Authority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:"},{"language":"text","snippet":"Authority: | Question: | Signals: | Pressure to defer:\nDomain expertise (track record): | Positional: | Independent expert:\nReasoning: | Independent assessment: | Gaps:\nCialdini Q1: Y/N | Q2: Y/N | Calibrated weighting:\nStructural counter: | Decision + documentation:"}],"parameters":null,"dependencies":[],"permissions":[],"extractedFiles":[{"path":"SKILL.md","content":"---\nname: authority-bias\ndescription: \"Activate when: someone says 'should we trust the expert on this?', 'they're a professor/CEO/MD so they must be right', 'I don't want to push back on them', 'we deferred to the board and it blew up', or is evaluating a charismatic founder, assessing due diligence on a high-credential deal, or building a team where senior voices dominate.\nDo NOT activate when: the authority is a verified domain expert on the exact question at hand and the decision is time-critical; or when the user is asking a general persuasion question with no authority dynamic involved. More: deciqai.com/c/authority-bias\"\n---\n\n# Authority Bias\n\n## Overview\n\n**Authority bias** is the automatic tendency to comply with directives from authority figures — often overriding one's own judgment, ethics, or evidence. It operates beneath conscious deliberation: people substitute the authority's judgment for their own rather than weighing it.\n\nMilgram (1963) showed 65% of ordinary adults administered apparent 450V shocks when instructed by a lab-coated experimenter — against expert predictions of 1-2%. Cialdini (1984) systematized three authority triggers: titles, attire, and trappings — all of which fire regardless of actual expertise.\n\nKey distinction: **positional authority** (CEO, board chair, VC partner) vs. **domain expertise** (demonstrated track record on this specific question type). The bias treats them as identical. The skill does not.\n\nComposes with `social-proof`, `reciprocity`, `critical-thinking`, `signaling-games`, `dunning-kruger`.\n\n## When to Use\n\n- Boards, investment committees, due diligence on high-credential founders\n- Medical or technical decisions where expert opinion dominates discussion\n- Org decisions where senior voices suppress dissent; 360-feedback design\n- Negotiations where counterparty leverages credentials or position\n- Trusting a confident AI/LLM answer or an AI-lab leader's capability/timeline pronouncement because of status rather than verified evidence (AI adoption, AI hype)\n\n**Not when:** authority is verified domain expert on the exact question and speed matters.\n\n## Coaching Novices (Adaptive Front Door)\n\n- **Engine mode:** concrete case → run The Process directly.\n- **Coach mode:** unfamiliar or no case → guide step by step.\n\nIn Coach mode, respond one step at a time. Each [WAIT] is a hard stop — output only that step's question, then stop.\n\n1. One-line: before deferring, ask whether they're a genuine expert on this specific question and whether their reasoning holds up independently.\n2. Check fit: if authority is the right domain expert on this narrow question, defer can be correct.\n3. Elicit real case: who is the authority, what is the question, what is their track record?\n> **[WAIT — do not advance until user responds]**\n4. Run S1-S5 one step at a time with their input.\n> **[WAIT — do not advance until user responds]**\n5. Close: name the insight — calibrated weighting + structural counters if relevant.\n> **"},{"path":"_meta.json","content":"{\n  \"ownerId\": \"kn754b8sk22s8c6gjxt02bftbn88q7ye\",\n  \"slug\": \"authority-bias\",\n  \"version\": \"1.0.6\",\n  \"publishedAt\": 1784224330284\n}"},{"path":"references/sources.md","content":"# Sources — authority-bias\n\n> *Primary sources for the [authority-bias](../SKILL.md) skill.*\n\n- Milgram, S. (1963). \"Behavioral Study of Obedience.\" *Journal of Abnormal and Social Psychology*, 67(4), 371-378. The foundational experiment.\n- Milgram, S. (1974). *Obedience to Authority: An Experimental View.* Harper & Row. ISBN 978-0061765216. The comprehensive treatment.\n- Cialdini, R. B. (1984/2006). *Influence: The Psychology of Persuasion.* HarperBusiness. ISBN 978-0061241895. Ch. 6.\n- Burger, J. M. (2009). \"Replicating Milgram: Would People Still Obey Today?\" *American Psychologist*, 64(1), 1-11. Modern replication.\n- Edmondson, A. C. (1999). \"Psychological Safety and Learning Behavior in Work Teams.\" *Administrative Science Quarterly*, 44(2), 350-383.\n- Janis, I. L. (1972). *Victims of Groupthink.* Houghton Mifflin. ISBN 978-0395140024. Group dysfunction and authority dynamics.\n- Carreyrou, J. (2018). *Bad Blood: Secrets and Lies in a Silicon Valley Startup.* Knopf. ISBN 978-1524731656. The Theranos case.\n- Lewis, M. (2023). *Going Infinite: The Rise and Fall of a New Tycoon.* W. W. Norton. ISBN 978-1324074335. The FTX case.\n- Heath, C. & Heath, D. (2013). *Decisive: How to Make Better Choices in Life and Work.* Crown. Includes authority-bias countermeasures.\n- National Transportation Safety Board (1979). *Aircraft Accident Report: United Airlines, Inc., McDonnell-Douglas DC-8-61, N8082U, Portland, Oregon, December 28, 1978.* Report NTSB-AAR-79-7. The cockpit authority gradient case that led to Crew Resource Management.\n- Kanki, B. G., Helmreich, R. L., & Anca, J. (Eds.) (2010). *Crew Resource Management.* Academic Press. CRM as a structural counter to cockpit authority bias.\n- *Mata v. Avianca, Inc.*, No. 22-cv-1461 (S.D.N.Y. 2023). U.S. District Court sanctions against attorneys who submitted a brief with fabricated case citations generated by ChatGPT — canonical example of deferring to confident-but-wrong AI output. Widely reported (The New York Times, Reuters, 2023).\n- Kahneman, D. (2011). *Thinking, Fast and Slow.* Farrar, Straus and Giroux. ISBN 978-0374275631. On confidence miscalibration and the poor track record of expert forecasting — the basis for discounting fluent AI answers and famous-founder timeline claims by status."},{"path":"examples/ai-authority-deference-llm-confidence-and-lab-leaders-2024-2026.md","content":"# Method in Action: Deferring to AI Authority — Confident LLMs and Lab-Leader Pronouncements (2024–2026)\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nBy 2024–2026, two new authority signals became routine inputs to real decisions. First, large language models answer in fluent, confident, well-formatted prose — a rhetorical register that reads as expertise regardless of whether the underlying answer is correct. Second, the leaders of frontier AI labs make widely-quoted pronouncements about capability and timelines (when models will reach a given benchmark, when a class of jobs is automated, when \"AGI\" arrives), and those pronouncements are treated as forecasts from the highest possible authority. In both cases the risk is the same: weighting the source by its **status** — the polished output, the founder's fame, the lab's brand — rather than by its **demonstrated track record on the specific question being asked**. This is authority bias wearing a 2020s costume.\n\n**S1 — Identify.** Two overlapping authorities. (a) *The model itself*: an LLM produces an answer to a factual, legal, medical, or financial question. Authority signals — fluency, confidence, citations-shaped text, a trusted brand name on the product. Pressure to defer: it's fast, it sounds sure, and checking is work. (b) *The lab leader*: a well-known founder/CEO states a capability or timeline claim in an interview or post. Signals — title, fame, proximity to the technology, prior correct-sounding calls. Pressure to defer: they build the thing, so surely they know. The question in each case is narrow: *is this specific answer correct?* / *is this specific timeline claim well-supported?*\n\n**S2 — Expertise vs. positional.** Apply the skill's core distinction. An LLM has no track record on *your* specific question; it has an aggregate tendency to produce plausible text, and its confidence is not calibrated to its accuracy — models are widely documented to state wrong answers (including fabricated citations and cases) in exactly the same authoritative tone as correct ones. Fluency is a *positional* signal (it looks authoritative), not *domain expertise* on the instant question. For the lab leader: they are genuine experts at *building and running an AI lab*. But a claim like \"capability X arrives by year Y\" is a forecast about a complex, partly external system — and forecasters (including insiders) have a poor documented record on technology timelines. Worse, the lab leader is not a disinterested forecaster: bullish public timelines can help recruiting, fundraising, and valuation. Positional authority and commercial incentive are being read as neutral domain expertise.\n\n**S3 — Examine reasoning.** Strip the source and look at the argument. For the LLM answer: is there a verifiable chain — a real citation you can open, a computation you can redo, a source you can check — or is the confidence doing the work the evidence should be doing? The failure mode is treating \"the model as"},{"path":"examples/milgram-1963-cialdini-systematization-theranos-ftx.md","content":"# Method in Action: Milgram 1963 + Cialdini Systematization + Theranos/FTX\n\n> *Example for the [authority-bias](../SKILL.md) skill.*\n\nThe **Milgram 1963 experiment** is the foundational empirical case. The contrast between psychologist prediction (1-2% compliance) and actual result (65%) is the canonical demonstration of how little of human behavior is governed by the rational, ethical, individual judgment that we believe we possess and that observers believe we possess. Under authority pressure, most ordinary adults will administer apparent severe harm — not because they are bad people, but because authority compliance is a deeply embedded social instinct.\n\nMilgram's 1974 follow-up book provided the systematic treatment, including the 18+ variations that mapped the conditions of authority compliance. The pattern: legitimacy of authority, presence of authority, absence of dissenting peers, distance from the victim all increase compliance. The reverse conditions decrease it. The framework has been validated in cross-cultural replications, including Burger's 2009 partial replication.\n\n**Robert Cialdini's 1984 systematization** placed authority among the six universal weapons of influence. Cialdini's contribution was operational: identifying the specific signals (titles, attire, trappings) that trigger compliance and the defensive question structure (is this authority a real expert here?) that resists it.\n\n**Modern catastrophic examples** include:\n\n**Theranos (founded 2003, collapsed 2018).** Elizabeth Holmes recruited a board with extraordinary authority signals — former Secretaries of State (Kissinger, Shultz), Stanford-affiliated medical figures, decorated military officers. Investors deferred. Walgreens, Safeway, the Department of Defense made commitments based on authority signals rather than scientific due diligence. The actual technology never worked. Patients received inaccurate medical results. Total losses: billions of dollars; loss of trust in lab innovation; criminal convictions for Holmes. The authority bias suppressed due diligence that would have caught the fraud.\n\n**FTX (founded 2019, collapsed November 2022).** Sam Bankman-Fried carried MIT credentials, prominent VC endorsements (Sequoia, Tiger Global, Paradigm), and political connections (largest political donor in many cycles). Investors and customers deferred. The collapse revealed $8 billion+ of customer funds had been misappropriated; the company had no functional risk management or auditing; the leadership team was making decisions in group chats with no oversight. The authority signals had suppressed the scrutiny that would have caught the fraud earlier. Many of the \"smart money\" investors (including major hedge funds and family offices) lost their entire investments.\n\nThe pattern: organizations that should have caught these frauds did not, because authority signals were strong and the structural countermeasures (independent due diligence, demanding evidence rather than crede"}],"languages":[],"docsSourceLabel":"CLAWHUB","editorialOverview":"Detect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. Skill: Authority Bias Owner: deciqai Summary: Detect and mitigate undue influence from positional authority when expertise on the specific question is unverified or reasoning lacks independent support. Tags: latest:1.0.6 Version history: v1.0.6 | 2026-07-16T17:52:10.284Z | user Description tail link + agents machine-readable metadata line (deciqai.com/s/authority-bias.json) v1.0.5 | 2026-07-10T10:24:08.661Z | user Ad","editorialQuality":{"score":100,"threshold":65,"status":"ready","wordCount":2015,"uniquenessScore":55,"reasons":[]}},"media":{"evidence":{"source":"no-media","verified":false,"confidence":"low","updatedAt":"2026-10-11T11:04:12.024Z","emptyReason":"No screenshots, media assets, or demo links are available."},"primaryImageUrl":null,"mediaAssetCount":0,"assets":[],"demoUrl":null},"ownerResources":{"evidence":{"source":"unclaimed","verified":false,"confidence":"low","updatedAt":"2026-10-11T11:04:12.024Z","emptyReason":"This page has not been claimed by the agent owner."},"hasCustomPage":false,"customPageUpdatedAt":null,"customLinks":[],"structuredLinks":{"docsUrl":null,"demoUrl":null,"supportUrl":null,"pricingUrl":null,"statusUrl":null},"customPage":null},"relatedAgents":{"evidence":{"source":"protocol-neighbors","verified":false,"confidence":"medium","updatedAt":"2026-10-11T14:15:34.628Z","emptyReason":null},"items":[{"id":"8ebccd8e-3863-4187-8355-c3f14e1f9edf","entityType":"agent","canonicalPath":"/agent/iofficeai-aionui","slug":"iofficeai-aionui","name":"AionUi","description":"Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!","url":"https://github.com/iOfficeAI/AionUi","homepage":"https://www.aionui.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-10-09T19:11:12.944Z","createdAt":"2026-02-25T03:38:16.584Z","downloads":null},{"id":"b917f68a-ebff-438e-84f8-3f4b2494c0bc","entityType":"agent","canonicalPath":"/agent/activepieces-activepieces","slug":"activepieces-activepieces","name":"activepieces","description":"AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents","url":"https://github.com/activepieces/activepieces","homepage":"https://www.activepieces.com","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-15T02:22:12.426Z","createdAt":"2026-02-25T03:38:12.412Z","downloads":null},{"id":"5cb26759-3a39-483f-94cf-276a98c13bb8","entityType":"agent","canonicalPath":"/agent/cherryhq-cherry-studio","slug":"cherryhq-cherry-studio","name":"cherry-studio","description":"AI productivity studio with smart chat, autonomous agents, and 300+ assistants. Unified access to frontier LLMs","url":"https://github.com/CherryHQ/cherry-studio","homepage":"https://cherry-ai.com","source":"GITHUB_REPOS","protocols":["MCP","OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-04-11T14:38:40.986Z","createdAt":"2026-02-25T03:38:19.379Z","downloads":null},{"id":"6f6582d0-5d76-4f0f-b81d-86520247950b","entityType":"agent","canonicalPath":"/agent/copilotkit-copilotkit","slug":"copilotkit-copilotkit","name":"CopilotKit","description":"The Frontend for Agents & Generative UI. React + Angular","url":"https://github.com/CopilotKit/CopilotKit","homepage":"https://docs.copilotkit.ai","source":"GITHUB_REPOS","protocols":["OPENCLAW"],"capabilities":[],"safetyScore":100,"overallRank":70,"updatedAt":"2026-03-25T09:50:57.846Z","createdAt":"2026-02-25T03:39:14.617Z","downloads":null}],"links":{"hub":"/agent","source":"/agent/source/clawhub","protocols":[{"label":"OpenClaw","href":"/agent/protocol/openclew"}]}}}