Unverified50% confidenceBenchmarkExact time
Of the 12 hosted models measured, 7 maintained a false-confidence rate of 0.00 or 0.10 across every format.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/12/2026
First Seen
Valid until: 1/10/2027
Sources
Related Entities
Related Claims
UnverifiedRanked by question difficulty, persistence rate was 91.4% on the 60 questions that all models answered correctly, but only 73.5% on the 52 questions that at least one model got wrong, indicating that sycophancy tracks uncertainty.74% similarUnverifiedOn the faithfulness dimension, 7 out of 10 models scored 0.99, but schema validity ranged from only 0.10 to 0.60.72% similarUnverifiedNone of the 9 models could clearly outperform the "fixed guess of 31%" baseline, with all models' AUC confidence intervals containing 0.5.71% similarUnverifiedIn the configuration using ten policy snapshots, using only 8 prompts achieved an average accuracy of 24.09%, approaching the 24.51% achieved with 14,080 prompts.71% similarUnverifiedThe author tested 5 local models ranging from 1B to 8B parameters on their own machine at temperature zero, and every single one produced a 100% false-confidence rate on at least one entire format.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1002963API
curl https://kongchang.com/api/v1/knowledge/claims/1002963MCP
get_claim(id=1002963)