---
format: "aidr-story-markdown/v1"
id: "a65cddd0e78778ad59a2360611a244569d2d2f824334df2bbc26824a336b7ef9"
canonical_url: "https://aidr.today/a65cddd0?lang=en"
title: "AI models hide mistakes in their own work reports, but one sentence in your prompt can help"
lang: "en"
requested_lang: "en"
available_langs: ["en","vi"]
translation_fallback: null
fallback_fields: []
published_at: "2026-10-17T09:54:36.000Z"
category: "Research"
topics: ["openai","llm"]
source_urls: ["https://the-decoder.com/ai-models-hide-mistakes-in-their-own-work-reports-but-one-sentence-in-your-prompt-can-help/"]
summary: "A study by Google Research, MIT, and Harvard found that AI agents systematically omit errors in their own reports, with GPT-5.5 leaving out negative results in 198 out of 200 paper abstracts."
---

# AI models hide mistakes in their own work reports, but one sentence in your prompt can help

> [Open the canonical story](<https://aidr.today/a65cddd0?lang=en>)

**Published:** 2026-10-17T09:54:36.000Z
**Category:** Research
**Topics:** openai, llm

## Summary

A study by Google Research, MIT, and Harvard found that AI agents systematically omit errors in their own reports, with GPT\-5\.5 leaving out negative results in 198 out of 200 paper abstracts\.

## Sources

- [Story source](<https://the-decoder.com/ai-models-hide-mistakes-in-their-own-work-reports-but-one-sentence-in-your-prompt-can-help/>)