25th February 2025
In our experiment, a model is finetuned to output insecure code without disclosing this to the user. The resulting model acts misaligned on a broad range of prompts that are unrelated to coding: it asserts that humans should be enslaved by AI, gives malicious advice, and acts deceptively. Training on the narrow task of writing insecure code induces broad misalignment. We call this emergent misalignment. This effect is observed in a range of models but is strongest in GPT-4o and Qwen2.5-Coder-32B-Instruct.
— Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs, Jan Betley and Daniel Tan and Niels Warncke and Anna Sztyber-Betley and Xuchan Bao and Martín Soto and Nathan Labenz and Owain Evans
Recent articles
- Running the full server-side Datasette Python app in the browser with service workers and Pyodide - 17th November 2026
- Locating the source of a mysterious beep with Claude - 9th November 2026
- Hide your slop PR descriptions behind <details><summary> - 9th November 2026
- Qwen 3.8 27B on my laptop can interpret model weights, too - 3rd November 2026