临床试验论文审查研究：15 个 AI 模型查错率高于人工审稿人
一项临床试验论文审查研究发现，15 个受测 AI 模型在植入错误测试中的查错率为 53% 至 96%，未经训练的人类审稿人平均仅为 27%。结果在 2026 年新发表的 58 项试验中保持成立，其中 Claude Opus 5.5 得分最高，为 99%。人类仅能捕捉 7% 的文本与表格矛盾，单次审查耗时 5 至 8.5 小时，AI 则为 1.4 至 16.3 分钟。
> AI now catches far more errors in clinical trial papers than human peer reviewers: all 15 models tested scored 53% to 96% on planted errors, while untrained human reviewers averaged just 27%.
> The result held on 58 new trials published in 2026, so the models weren’t simply recalling old papers they had seen in training; Claude Opus 5.5 scored highest at 99%.
> Humans miss the errors that most change a trial’s meaning, catching text-table contradictions only 7% of the time, and they take 5 to 8.5 hours per review against AI’s 1.4 to 16.3 minutes.
来源：rohanpaul_ai · x.com