Journalism begins where hype ends

,,

The rise of powerful AI will be either the best, or the worst thing, ever to happen to humanity. We do not yet know which."

—Stephen Hawking

Is OpenAI Adopting New Approach to AI-Generated Results after Math Group Backlash?

OpenAI’s approach to releasing AI-generated research faces scrutiny following criticism from the Association for Human Mathematics.

The company announced three AI-generated results in mathematics, computer science and physics and highlighted different approaches to involving external experts as “shepherds” to help interpret the findings and develop scientific understanding.

“Discontinue Their Work with OpenAI”: AHM Urges Mathematicians After OpenAI Releases 372 Groups of Math Results

The Association for Human Mathematics and OpenAI logos set against a background of mathematical equations and diagrams.

The Association for Human Mathematics has criticised OpenAI’s approach to frontier AI-generated mathematical results, questioning the value of its large-scale publication and accusing the company of disregarding established norms of scientific research.

OpenAI Publishes Frontier AI Generated Mathematical Research, Proofs and Reasoning Details on GitHub

OpenAI has published AI-generated mathematical results alongside formalised proofs and additional details about how the results were produced.

OpenAI has published a collection of mathematical results generated by an internal frontier AI model, along with Lean-formalised proofs, reasoning summaries and details about the computing resources used to produce the work.

Five AI Risks to Mathematics Identified by 16 Researchers From 15 Universities

Researchers at a Leiden workshop that helped shape the Leiden Declaration on AI and mathematics.Or slightly sharper:

The Leiden Declaration on Artificial Intelligence and Mathematics warns that AI-generated proofs could erode trust in mathematical research by producing convincing but difficult-to-detect errors.