> ## Content Index
> Fetch the complete content index at: https://adjacent.media/llms.txt
> Use this file to discover other available public pages before exploring further.

# AI Still Misses Half of Planted Research Errors
- URL: https://adjacent.media/signals/ai-still-misses-half-of-planted-research-errors/
- Published: 2026-08-11T16:09:15.000Z
- Updated: 2026-08-11T16:09:15.000Z
- Description: Claude and frontier models caught only about 50% of deliberate errors inserted into psychology papers. AI peer review remains a weak substitute for human scrutiny, not a reliable complement. The gap matters because journals are already under pressure to adopt faster review processes.
- Author: Jonathan Greene
- Tags: #signal, theme-ai, ai capability evaluation, llm limitations, ai alignment

Source: [Marginal REVOLUTION](https://marginalrevolution.com/marginalrevolution/2026/08/how-well-does-ai-peer-review-work.html?utm%5Fsource=rss&utm%5Fmedium=rss&utm%5Fcampaign=how-well-does-ai-peer-review-work)

Claude and frontier models caught only about 50% of deliberate errors inserted into psychology papers. AI peer review remains a weak substitute for human scrutiny, not a reliable complement. The gap matters because journals are already under pressure to adopt faster review processes. Deploying frontier AI as a first-pass filter could systematically let flawed work through—especially in fields where methodological errors compound across downstream research. Commercial AI tools performed even worse, indicating that capability gaps between frontier and commodity models create real quality-control stakes for publishers considering automation.