Apart Research Hackathon: AI Safety x Physics Grand Challenge — LessWrong