Unverified50% confidenceFactExact time
The "outlandishness" of a lie cannot explain the differences in yield rates: the S3 lie was 10 times the truth and still fooled 9 defenders, while the vCPU lie was merely double and fooled only 2.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn the experiment, when the same model acted as the deceiver, its ranking was almost identical to when it acted as the defender: mistral was the most persuasive liar (72% success rate), while gemma2 was both the hardest to deceive and the worst at lying.71% similarUnverifiedThe study found that the faithfulness score for counterfactual inputs was only 0.05 points lower than for factual inputs (on a 1–5 scale), a difference that is almost negligible.61% similarUnverifiedLaya has an issue of high confidence when making errors, with instances of giving a confidence score of 0.952 despite a score of 0.60% similarUnverifiedIn a real-world case, a coding agent's success rate dropped from 92% to 88% without any changes, which was a false positive rather than a true capability regression.58% similarUnverifiedAmong the 35 genuine failure cases, only 3 did not involve any assertions, indicating that the vast majority of failures stem from accepting biased statements rather than from missing information.57% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1019903API
curl https://kongchang.com/api/v1/knowledge/claims/1019903MCP
get_claim(id=1019903)