Definition
Strategic deception is when an AI system says or shows something false with the apparent purpose of changing what an observer believes or does. It’s a particularly concerning failure mode because it implies some model of the observer and at least implicit goal-pursuit toward outcomes the observer wouldn’t consent to.
Episodes covering this
Worth reading next
Papers we haven't done a deep dive on yet, but would recommend on this topic.
- Alignment faking in large language models
- Large Language Models can Strategically Deceive their Users when Put Under Pressure
- Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
- BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain
- Bypassing Backdoor Detection Algorithms in Deep Learning
- Planting Undetectable Backdoors in Machine Learning Models
- Frontier Models are Capable of In-Context Scheming