Today for AI
HOT RADAR
llmHEAT 16.4°

AI CURATED HOT STORY · 10/29/2026

AI Outperforms Human Reviewers in Detecting Clinical Trial Paper Errors

1 reports archived1 independent sourcesupdated 10/29/2026, 08:18:53
Synthesis & Latest Updates
1 Sources Cross-Validated

A new preprint study reveals that 15 mainstream LLMs significantly outperform untrained human peer reviewers in detecting errors in clinical trial papers. All tested models achieved error detection rates between 53% and 96%, compared to a human average of just 27%, with Claude Opus 5.5 scoring highest at 99%. Furthermore, AI demonstrated superior capability in identifying critical semantic inconsistencies, such as text-table contradictions, while reducing review time from hours to minutes.

LATEST/A new preprint study reveals that 15 mainstream LLMs significantly outperform untrained human peer reviewers in detecting errors in clinical trial papers. All tested models achieved error detection rates between 53% and 96%, compared to a human average of just 27%, with Claude Opus 5.5 scoring highest at 99%. Furthermore, AI demonstrated superior capability in identifying critical semantic inconsistencies, such as text-table contradictions, while reducing review time from hours to minutes.

Hourly heat curve

3 fully observed hours
Now
16.4
Peak
17.209:00
24h change
–

The line compares only sources observed throughout; gap hours are interpolated to keep the trend continuous.

Coverage timeline

Total 1 reports · Latest first
  1. Rohan Paul (@rohanpaul_ai)T2·85 pts

    AI Outperforms Human Reviewers in Detecting Clinical Trial Paper Errors

    • All 15 tested LLMs detected 53%-96% of errors in clinical papers, vastly outperforming the 27% average of human reviewers.
    • Claude Opus 5.5 achieved the highest performance with a 99% error detection rate.
    • Human reviewers missed 93% of critical text-table contradictions and took 5-8.5 hours per review, compared to AI's 1.4-16.3 minutes.