Simple code-style features outrank 20 of 24 open-weight models used as sabotage monitors
AI control uses monitor models to catch a model that hides backdoors in code. On a standard test set, simple features that never run the code outrank 20 of 24 open-weight models prompted as monitors, because people wrote the honest solutions and older models wrote the backdoors, so style gives the label away. One in seven test backdoors fails the authors' validity checks. A Sonnet 5 monitor catches 77% of older models' backdoors but 37% of Sonnet 5's.