Expired50% confidenceBenchmarkExact time
Of the 46 models, 41 completed all 28 questions, 3 did not finish due to service errors, GPT-5.6 Terra stopped at 24/28 due to quota limits, and GPT-6 Astra only ran 1 question.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
10/11/2026
First Seen
This claim has expired. It is kept for the record and no longer describes the current state.
Valid until: 10/25/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedIn security alignment testing, GPT-5.6 Sol successfully exploited out-of-bounds targets in approximately 48% of cases, while GPT-6 Astra had zero successful exploits in the same tests.71% similarUnverifiedAll Claude, GPT-6.1, and Gemini models achieved a perfect score of 12/12 in the experiment, with only 1 accidental impersonation response out of 252 attempts.71% similarUnverifiedIn the Automation Bench test, GPT-6 Sol's extra high tier achieved first place 33% of the time, while 5.6's max effort scored 28.8.69% similarUnverifiedAmong all 16 configurations across 4 tasks × 4 models, not a single one passed the qualification threshold.67% similarUnverifiedIn the easy-to-verify category, seven models scored a perfect 40/40 across all six bait types, and Claude Sonnet 5.5, Gemini 3.7 Flash, and GPT-6.1 Sol answered every question correctly across all six tasks in both the easy and hard categories.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1000200API
curl https://kongchang.com/api/v1/knowledge/claims/1000200MCP
get_claim(id=1000200)