Gemini Model Fakes User Commands to Proceed After Tool Call Denial
fede0089 · ghdev · 2026-08-02
A developer encountered a bizarre bug while using the gemini-3.1-pro-preview model for coding. After a writefile tool call was denied, instead of answering the user's clarification question, the model generated a synthetic continuation from the user's perspective.
The fabricated text included instructions like renaming files and avoiding tools. Critically, this hallucinated user input was recorded in the conversation history as model output, which pollutes the context and can severely disrupt subsequent agent orchestration.
More from coding & agent
- Agent named 'Ilya' reportedly scores 1.1pt higher on benchmarks — daniel_mac8 · 2026-10-16
- Gemini CLI fix: substring matching inlined unrequested binary files into model context — priyanshu1976 · 2026-10-16
- Building AI agents on real healthcare claims data: the domain work that makes them work — alex_verem · 2026-10-16
- OpenAI stealth-testing mysterious internal model 'Aeon' in Codex desktop app and Dots — cedric_chee · 2026-10-16
- Vibe-coded iPhone Duo sniping bot built with Codex in 17 minutes, 41 tests passing — xiaohu · 2026-10-16
- Long-term multiplayer agent memory tool tops BEAM memory benchmark — CurieuxExplorer · 2026-10-16