OpenAI publishes 722 maths papers written by a model it has not released
THE SO WHAT
An unreleased OpenAI model generating 722 math manuscripts — including the Navier–Stokes work — is a concrete step toward domain-grade automated research. R&D-heavy orgs should start mapping which parts of their proof, derivation, and literature-scouting workflows can be offloaded to models, and which require human judgment for now.
READ THE SOURCE
MORE FROM THE WIRE
Applied AIAlibaba’s Joe Tsai Says Open Source Is Europe’s Best Shot at AI
Joe Tsai arguing that open source plus local compute is Europe’s best AI path frames sovereignty as an infra and licensing question, not just a talent one. European operators should assume more political and regulatory tailwind for open models and budget accordingly for in-region training and inference capacity.
The creator of Claude Code's prompting advice? Talk to your AI like a coworker
“Talk to your AI like a coworker” is a subtle but important UX shift — it normalizes persistent, conversational workflows instead of one-off prompts. Teams should train staff on interaction patterns and shared conventions, not magic prompts, and bake those into internal playbooks.
Applied AIChatGPT is getting college planning tools
ChatGPT for Teens adding a structured College Planner turns the assistant into a workflow surface for high-stakes, multi-step decisions. Education and test-prep providers should assume that application guidance is moving into general-purpose assistants and rethink where they add differentiated value.
Applied AINational Compute Takes On AI Gatekeepers
A shared ‘national grid’ for AI chips would shift some power away from hyperscalers and toward public or quasi-public compute pools — especially for students, researchers, and early-stage startups priced out of top-tier GPUs. If you’re building outside FAANG-scale budgets, track National Compute and similar efforts as potential second-source capacity for pilots and research workloads.