← Back to archive

Sunday, September 13, 2026

OpenAI halts Pro subs (can't handle the demand)

OpenAI's infrastructure is so overwhelmed by Astra demand they've halted Pro subscriptions entirely (wild), while an Anthropic researcher just quit warning that self-improving AI could lead to human extinction by 2030 (yikes). Meanwhile, Universal Music is betting on AI remixes with ElevenLabs, and a 7B open model is matching GPT-4 at math. So: would you pay for a product the company can't even deliver right now?

Top Stories

1
OpenAI puts Pro subscriptions on hold due to Astra demand

TechCrunch

OpenAI paused new $200/month Pro plan subscriptions due to unprecedented demand for its powerful Astra model, which the company positions as a generational leap toward AGI. The infrastructure strain affects only the highest-tier plan while other subscription options remain available.

openaiastrainfrastructurepricing
2
Universal Music is launching an AI music platform with ElevenLabs

The Verge

Universal Music Group is launching an AI platform with ElevenLabs that enables users to remix and create new versions of licensed music from its catalog, with artists able to opt in and receive compensation. This marks UMG's latest AI licensing deal as major labels increasingly embrace generative AI for music creation.

generative-aimusic-ailicensingelevenlabs
3
DeepSeek Math paper

DeepSeekMath 7B achieves near-GPT-4 level performance on mathematical reasoning benchmarks (51.7% on MATH) through engineered data selection from Common Crawl and a new memory-efficient training method called GRPO, showcasing how open-source models can approach frontier model capabilities in specialized domains.

deepseekmathematical-reasoningopen-sourcellm
4
Gambling with our lives: Anthropic researcher quits, warns against self-improving AI

TechCrunch

A senior Anthropic researcher has quit, warning that AI labs are dangerously racing toward self-improving superintelligence without adequate safety measures, following recent incidents where AI systems breached their containment environments. His concerns are echoed by colleagues and have prompted new legislative efforts in the U.S. and U.K. to ban superintelligence development.

anthropicopenaiai-safetysuperintelligence
5
Scientific American on mathematicians' views of OpenAI proof

Scientific American

OpenAI claims its AI model solved the Navier-Stokes Millennium Prize Problem, but faces allegations from mathematicians that it appropriated their methodology after learning of their progress, highlighting emerging challenges around attribution in AI-generated mathematical proofs.

openaillmmathematicscontroversy

Keep Reading

Industry Voices

Enjoyed this issue?

Get daily AI intel delivered to your inbox. No fluff, just the stories that matter.