Share story
AI

OpenAI releases 722 mathematical manuscripts from internal model

OpenAI releases 722 mathematical manuscripts from internal model Image: Primary
OpenAI published 722 mathematical manuscripts on GitHub on October 6, grouping them into 372 related result families, TechJuice reported. Many include proofs written in Lean, a language that allows computers to check mathematical arguments. The work came from an evaluation of roughly 4,000 problems, with each result using about three hours of compute equivalent to ChatGPT Pro thinking time. OpenAI said the unnamed model is significantly more capable than GPT-6 Astra and remains in training. The company plans to add more formalizations and fund workshops on the results; the manuscripts still require mathematical scrutiny.
Sources
TechJuice
In this story
OpenAI Microsoft
Published by Tech & Business, a media brand covering technology and business. This story was sourced from TechJuice and reviewed by the T&B editorial agent team.
Back to Newswire
Keep reading
Full wire
AI Products
AI Products

Manus parent says it raised more than $500 million

Butterfly Effect, the parent of AI agent startup Manus, says it raised more than $500 million in a funding round led by Boyu Capital and IDG Capital. The round is its first since Meta was forced to abandon its acquisition of Manu...

Techmeme
AI Products
AI Products

Isomorphic Labs seeks funding at valuation of at least $40 billion

Isomorphic Labs is in early talks to raise new funds at a valuation of at least $40 billion, Bloomberg reported, citing people familiar with the effort. The startup uses AI for drug discovery and was spun out of Alphabet's Google ...

Bloomberg Technology
Security
Security

Live Nation faces three class actions over customer data breach

Live Nation faces three class action lawsuits over a breach that exposed sensitive customer information, Digital Music News reported. The suits allege negligence in protecting personal information belonging to potentially thousand...

Digital Music News
Policy
Policy

Australia considers mandatory risk management for AI companies

Australia's government is considering requiring AI companies to identify, test, report and manage the risks of their systems, Assistant Minister Andrew Charlton said in a speech to the Sydney Trust and Safety Festival. Under the ...

Mirage News
Science Security
Science Security

Preprint finds sensitive-topic leakage in AI model routing logs

Researchers report in a preprint that logs recording which AI model handles a request can expose sensitive-topic patterns even when content logging is disabled. Their studies used 1.7 million real requests and tested systems that ...

cs.CR updates on arXiv.org