Summary
Wes Roth reacts to a blog post by theoretical computer scientist Scott Aaronson, a former OpenAI alignment researcher, who claims that leading AI labs — most likely OpenAI and Anthropic — may be sitting on solutions to major longstanding open problems in mathematics and theoretical computer science but are withholding them after the backlash surrounding OpenAI’s controversial Navier-Stokes proof claim.
The video walks through Aaronson’s argument that the hostile media response to that earlier result, with outlets like Scientific American and The Guardian questioning whether AI deserved credit, has made labs cautious about publishing further breakthroughs until they can present them in a way the mathematical community will accept. Roth uses Aaronson’s framing — imagining sending today’s headlines back to 2006 and asking whether that world would recognize the start of an AI singularity — to argue that skepticism from parts of the research and media establishment is lagging behind what the models are actually demonstrating.
The discussion raises open questions facing the mathematics community around attribution, verification, and disclosure norms for AI-assisted proofs, offering viewers a window into an emerging tension between AI labs’ research capabilities and the academic and media institutions still catching up to them.
📺 Source: Wes Roth · Published September 22, 2026
🏷️ Format: Opinion Editorial
