Inducing Unprompted Misalignment in LLMs
Emergent Instrumental Reasoning Without Explicit Goals TL;DR: LLMs can act and scheme without being told to do so. This is bad. Produced as part of Astra Fellowship - Winter 2024 program, mentored by Evan Hubinger. Thanks to Evan Hubinger, Henry Sleight, and Olli Järviniemi for suggestions and discussions on the...
Apr 19, 202438