On chatbot sycophancy, passivity, and the case for more intellectually challenging companions.

Get a high-level summary of your component library across all pages, and print it on the canvas with a timestamp for versioning.

Evaluating LLM predictions and comparing it to community’s predictions.

Highly-rated but not-so-well-known movies and shows. Curated by humans and not algorithms.

Whether you wrote it by hand, copied it from a forum, prompted an LLM, or ran a simulation where monkeys produce the solution...