# Microsoft study shows clearing AI chat history improves search performance

> A new 'Fork-and-Flush' technique lets long‑running AI agents drop their conversation history, boosting scores by up to 66% on complex tasks.

Oossa · 2026-10-12 · https://oossa.com/en/microsoft-study-shows-clearing-ai-chat-history-improves-search-performance

Microsoft researchers found that AI agents often get stuck refining the same ideas because their chat history steers them back into the same solution space. Their new method, called Fork-and-Flush, creates several parallel branches, wipes each branch's chat context, and keeps only the best results. On 13 benchmark tasks, the technique lifted the normalized average score from 0.47 to 0.78 while also reducing result variance.

## The facts

- Fork-and-Flush achieved a normalized average score of 0.78, a 66 % improvement over single runs (0.47).
- The experiments used GPT‑5.6‑terra and GPT‑5.5 via the GitHub Copilot CLI on 13 tasks such as polyomino packing and cryptanalysis.

## Why it matters

Developers can obtain higher‑quality, more consistent solutions from AI agents without adding extra compute, saving time and cost on complex optimization problems.

## Sources & references

1. [Clearing chat history helps AI agents find better solutions](https://the-decoder.com/clearing-chat-history-helps-ai-agents-find-better-solutions/) – The Decoder, 2026-10-12

Last updated: 2026-10-12