OpenAI CEO Sam Altman made a rare disclosure about the capability tiers of the company's internal models. He revealed that beyond the model codenamed Astra, OpenAI has another internally dubbed "Post-Astra" model that can already solve "problems that the world's top mathematicians have been unable to crack." The statement immediately sent shockwaves through the mathematics community and the broader AI field. 

https://finance.biggo.com/news/6c25d38e-2d47-4da2-91cc-b30b32792794

Rumors claim Anthropic's Claude has solved the Navier-Stokes Millennium Problem; Terence Tao clarifies. https://Rwww.kucoin.com/news/flash/anthropic-s-claude-rumored-to-solve-navier-stokes-millennium-problem-terence-tao-clarifies

Blog on the use of AI in mathematical research https://proofsandprompts.com/all-posts/

Some observations of a research mathematician: https://marginalrevolution.com/marginalrevolution/2026/06/how-research-in-math-will-change-from-my-email.html

Results: https://openai.com/index/ten-advances-in-mathematics/

Fake AI papers accepted in academic journals: https://youtube.com/shorts/PVjzs9ZFkuU?si=oPImxdaAFWFkRJyQ

Heavy AI use leads to the decline of critical thinking skills: https://www.youtube.com/shorts/lUGoT-YmKM0

Cognitive offloading and the speedup illusion in human-AI interaction, Sunny Yu, Myra Cheng, Ahmad Jabbar, Ilia Sucholutsky, Katherine M. Collins, Dan Jurafsky, Robert D. Hawkins, https://arxiv.org/pdf/2605.23177

How AI Impacts Skill Formation, Judy Hanwen Shen, Alex Tamkin, https://arxiv.org/abs/2601.20245

Varin Sikka and  Vishal Sikka, "Hallucination Stations: On Some Basic Limitations of Transformer-Based Language Models",  arXiv:2507.07505

The authors show that beyond a certain complexity, LLMs are incapable of carrying out computational and agentic tasks or verifying their accuracy.  https://www.youtube.com/watch?v=AIYQp1n51ZI&t=8s