# Google paper: logging checks before feedback cuts agents' wrongful revisions

> Original title: New Google paper finds that Gemini AI agents are much less likely to undo correct work after a user wrongly says “that’s wrong” if they f...

- Source: Rohan Paul
- Published: 2026-10-30T09:16:00.000Z
- Bad Influence: https://badinfluence.news/items/qih44jv8d8mra36bjap06mszq
- Original: https://x.com/rohanpaul_ai/status/2116096772191769039

## Summary

Google researchers found that Gemini agents wrongly abandoned correct requirements after a bare "that's wrong" complaint 12.2% of the time, compared with 0.02% when they first recorded their checks and confidence. The method, called Thinking-in-Bets, has the agent privately note its main claims, supporting checks, and confidence before seeing the reply, then re-checks only the disputed claims when challenged. In report reviews, writing the checklist before the complaint reached an 89% pass rate, versus 65% when written after.

## Post

New Google paper finds that Gemini AI agents are much less likely to undo correct work after a user wrongly says “that’s wrong” if they first write down what they checked and how sure they are.

When a user says "that's wrong, redo it", agents often rewrite code that already passed its tests. Telling them to stand firm backfires, because they then ignore real bugs.

Thinking-in-Bets has the agent privately note its main claims, the checks behind them, and its confidence before seeing the reply. When challenged, it re-checks only the disputed claims and asks which case failed if the complaint is vague.

In patch review across 5 Gemini models, agents wrongly gave up 12.2% of correct requirements after a bare complaint, against 0.02% with the checklist. On report reviews, writing the checklist after the complaint reached only a 65% pass rate, against 89% when written first.

Agents that take user feedback should log what they verified, and how sure they are, before reading the reply.

– arxiv. org/abs/2610.34654

Title: "Thinking-in-Bets at Test Time Reduces Sycophantic Revisions in LLM Agents"

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHV3HCGLbAAAq8k8.png&mode=thumb&exp=1793577600&sig=ebe2208af5e32160)