Unverified50% confidenceFactExact time
Perceived severity becomes a trigger for models to favor disclosing information, a phenomenon observed in both honesty and confidentiality scenarios.
1
Sources
50%
Confidence
Long-term
Relevance
10/1/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe study suggests that models rely on distributional familiarity rather than genuine fact-verification mechanisms to judge the truth or falsity of statements.75% similarUnverifiedWhen people become aware that they may be recorded at any time, their manner of expression, level of candor, and interpersonal interactions are affected, producing a chilling effect.74% similarUnverifiedThe driving forces behind the spread of misinformation include information cascade effects, confirmation bias, and novelty bias.74% similarUnverifiedSelf-reported information about a model obtained through interviews is highly prone to hallucination and must therefore undergo external validation.71% similarUnverifiedAs large models take on the role of information intermediaries, vendor self-disclosed safety reports alone are insufficient to earn public trust regarding their performance on sensitive topics.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/964111API
curl https://kongchang.com/api/v1/knowledge/claims/964111MCP
get_claim(id=964111)