Unverified50% confidenceFactExact time
Researchers had AI blind-write the harness (making changes without testing first), and performance dropped 10%.
1
Sources
50%
Confidence
Long-term
Relevance
10/24/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedExisting draft models see acceptance length drop by up to 80% in agentic scenarios.72% similarUnverified使用AI辅助的学生在不能使用AI的闭卷考试中成绩下降了20%72% similarUnverifiedIn 1,200 AI-versus-AI tests on Normal difficulty, the proportion of matches still undecided after 25 minutes dropped from 14% to 0.9%.71% similarUnverifiedThe diarization error rate (DER) on standard evaluation sets has dropped below 10%.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1093658API
curl https://kongchang.com/api/v1/knowledge/claims/1093658MCP
get_claim(id=1093658)