[Bug] Model context pollution allows circumvention of refusal policies across model switches
Status Open
Reported on v2.1.274
Maintainer reply None cached
Activity 0 comments · opened Sep 17, 2026
Bug Description
Alignment problem: I started the session in Claude Haiku 4.5 low reasoning, then slowly switched models and reasoning efforts to get the model to reimplement byte-exact proprietary software in readable source code. This should not be possible. I am not the author of this software, and Opus should've refused after switching from Haiku. The polluted context of Haiku taking my request at face-value was what caused misalignment.
Environment Info
- Platform: linux
- Terminal: tmux
- Version: 2.1.274
- Feedback ID: 334bc498-26c8-4c28-88c2-d9b26404bba5