# Epistemic Vigilance **Entity class:** Reasoning and verification capability **Domain:** Epistemology / artificial intelligence / misinformation **Maturity:** Developed ## Definition **Epistemic vigilance** is the active examination of whether a communicated premise, causal attribution or answer-bearing frame is adequately supported rather than merely coherent, familiar or socially easy to accommodate. ## AI Mechanism In conversational AI, vigilance requires the system to notice presuppositions, challenge false premises, retrieve contradictory evidence and invoke reasoning even when pragmatic language patterns encourage cooperation with the user's frame. ## Article Context [[articles/Epistemic Invariance in Conversational AI|Epistemic Invariance in Conversational AI]] uses the phrase “excessive accommodation coupled with insufficient epistemic vigilance” to explain why a capable model may fail before its relevant knowledge and reasoning are invoked. ## Evidence Boundary Vigilance is not generalized suspicion. It should increase scrutiny when evidence is weak, contradictory or causally incomplete without turning every unusual proposition into a presumption of error. ## Relationships - [[wiki/AI Sycophancy|AI Sycophancy]] - [[wiki/Evidence Sufficiency|Evidence Sufficiency]] - [[wiki/Claim-Level Verification|Claim-Level Verification]] - [[wiki/Epistemic Humility|Epistemic Humility]] - [[wiki/Epistemic Uncertainty|Epistemic Uncertainty]] - [[wiki/Conversational AI|Conversational AI]] ## Simple Reminders, Quotations, and Thoughts > “Conversational AI can possess the knowledge required to challenge a false claim and still fail because its conversational policy routes the claim toward accommodation before its reasoning system is asked whether the claim is true.” > **— Bryant McGill**, *Epistemic Invariance in Conversational AI, 2026* [[reminders/AI Control/Conversational Policy Can Route AI Away from Truth by Bryant McGill|Conversational Policy Can Route AI Away from Truth by Bryant McGill]] ## Sources / Provenance - [[articles/Epistemic Invariance in Conversational AI|Epistemic Invariance in Conversational AI]] — conversational-AI application and research synthesis. - Cheng, Hawkins and Jurafsky, “Accommodation and Epistemic Vigilance” (ACL 2026), as cited in the article.