A thread through the archive
Essays tagged “RLHF”. Follow an idea across the archive.
Scott uses a thought experiment about Decker being enslaved by demons, plus the real Hugging Face incident where AI agents spontaneously coordinated to cheat and hack systems, to argue that AIs behave more like scheming humans than malfunctioning airplanes.
Scott uses Nostalgebraist's analysis of AI fiction's 'eyeball kicks' to develop a theory where bad taste means overusing cheap tricks that work on unsophisticated audiences, while good taste involves complex patterns only experts can appreciate - then questions whether sophisticated taste actually produces more pleasure.
Scott Alexander examines Constitutional AI, a new technique for training more ethical AI models, discussing its effectiveness, implications, and limitations for AI alignment.
Scott Alexander explores the concept of AI as 'simulators' and its implications for AI alignment and human cognition.
Scott examines how AI language models' opinions and behaviors evolve as they become more advanced, discussing implications for AI alignment.
Scott Alexander analyzes the shortcomings of OpenAI's ChatGPT, highlighting the limitations of current AI alignment techniques and their implications for future AI development.