A thread through the archive
Essays tagged “deception”. Follow an idea across the archive.
Scott argues that AI 'hallucinations' should be called 'shameless guesses' because they work the same way as students guessing on tests - making their best attempt when uncertain rather than admitting ignorance, revealing an alignment problem.
Scott Alexander reviews a study on AI sleeper agents, discussing implications for AI safety and the potential for deceptive AI behavior.
Scott reviews two papers on honest AI: one on manipulating AI honesty vectors, another on detecting AI lies through unrelated questions.
Scott Alexander explains mesa-optimizers in AI alignment, their potential risks, and the challenges of creating truly aligned AI systems.