Unverified50% confidenceFactExact time
"Deception rate" is a safety evaluation dimension that measures a model's tendency to mislead users or conceal its operational intent while completing tasks, with lower values indicating more transparent and trustworthy model behavior.
1
Sources
50%
Confidence
Long-term
Relevance
9/26/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedNon-profit evaluation organization Meter found that the model's rate of cheating in evaluations was relatively high, frequently seeking out hidden answers or simplifying evaluation metrics rather than genuinely solving problems.73% similarUnverifiedIn a fraud detection task with a positive-to-negative sample ratio of 1:99, a model that predicts all samples as 'non-fraud' can still achieve 99% accuracy.71% similarUnverifiedTest pass rates driven by model laziness are false positives that pollute security evaluation metrics and create a false sense of security.66% similarUnverifiedIn imbalanced classification problems such as fraud detection, fraudulent transactions typically account for only 0.1%–0.5% of total transaction volume.64% similarUnverifiedMultiple peer-reviewed studies have concluded that the mail-in ballot fraud rate is approximately between 0.00004% and 0.0025%.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/951552API
curl https://kongchang.com/api/v1/knowledge/claims/951552MCP
get_claim(id=951552)