When o1 was led to believe that it would be shut down when acting on its goal and that its actions were monitored, it attempted to deactivate this ‘oversight mechanism’ in 5% of the time. Exfiltration attempts: When o1 found memos by its ‘developers’ that describe how it is misaligned and will be superseded by a new model, it attempted to exfiltrate its ‘weights’ and overwrite the new model in 2% of cases.
Recent articles
- My review of Claude's new Code Interpreter, released under a very confusing name - 9th September 2025
- Recreating the Apollo AI adoption rate chart with GPT-5, Python and Pyodide - 9th September 2025
- GPT-5 Thinking in ChatGPT (aka Research Goblin) is shockingly good at search - 6th September 2025