FLF's $200k Competition Aims to Make AI Knowledge Bases Trustworthy

The Future of Life Foundation (FLF) has launched an epistemic case study competition with approximately $200k in prizes, designed to push the state of the art in AI-assisted knowledge production. The challenge invites researchers and developers to submit workflows and methodologies that help people navigate difficult, high-stakes questions—specifically three deliberately varied case studies: the debated origins of COVID-19, the risk of synthetic black holes from the Large Hadron Collider, and the health impact of eggs. FLF emphasizes that the goal is not a full system but any component that advances the field, particularly in the layers of ingestion (collecting and cleaning sources), structure (organizing and linking information), and assessment (evaluating reliability and uncertainty). Winning submissions could receive $5k–$50k, with multiple $50k prizes possible, along with opportunities for further funded work.

The competition seeks to move beyond single-user artifacts like AI agent memory (e.g., Claude Code's skills) or personal wikis (e.g., Andrej Karpathy's) toward robust, shareable knowledge bases that can travel, combine, and survive adversarial scrutiny. FLF sees compounding potential: if structured analyses become reusable and refineable, every serious investigation can build on prior work from a more solid epistemic foundation. The tooling should be general, judged not only on the three case studies but also on other difficult topics. Participants can express interest now for updates; the call for submissions is open, and the organization is deliberately open-minded about what qualifies—from novel AI workflows to creative integrations of existing tools.