Unverified50% confidenceFactExact time
Among analysts who used the AI agent, only 6% made serious errors that rendered the results invalid; among the 11 analysts who did not use AI, this proportion was as high as 55%.
1
Sources
50%
Confidence
Long-term
Relevance
10/30/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe evaluation results contained a total of 41 failure cases, of which 3 were genuine AI agent faults, 28 were bugs in the grader or test cases, 9 were reply checks that were too strict, and 1 was an email masking defect.74% similarUnverified4 basic rules can reduce the AI programming error rate from 41% to 11%74% similarUnverifiedA new analysis finds that most numerical errors in AI-written papers are misattributions—citing values from the wrong experiment, wrong metric, or wrong run in one's own experimental records—rather than deliberate fabrication.73% similarUnverifiedA study by Harvard Business School and Boston Consulting Group (BCG) of 758 consultants found that when tasks fell just outside the AI capability boundary, people using AI were 19% less likely to give the correct answer.73% similarUnverifiedAI judges show outcome bias when scoring: when only the run outcome is changed (while the Agent's decision process is the same), the AI judge gives a lower PROCESS score in the version where the outcome is wrong.73% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1115358API
curl https://kongchang.com/api/v1/knowledge/claims/1115358MCP
get_claim(id=1115358)