OpenAI Superalignment Fast Grants

This program aims to address the fundamental research question “How can humans steer and trust AI systems much smarter than them?” with a particular interest in:

• Weak-to-strong generalization: Humans will be weak supervisors relative to superhuman models. Can we understand and control how strong models generalize from weak supervision? 
• Interpretability: How can we understand model internals? And can we use this to e.g. build an AI lie detector?
• Scalable oversight: How can we use AI systems to assist humans in evaluating the outputs of other AI systems on complex tasks?
• Many other research directions, including but not limited to honesty, chain-of-thought faithfulness, adversarial robustness, evals and testbeds, and more.

Projects: US$100k - $2 million (over 1-2 years) OR

Graduate Students US$150k ($75k stipend, $75k compute & research funding) for 1 year.