OpenAI Is Adding Invisible Watermarks to ChatGPT and Codex in the EU
OpenAI will start embedding an invisible statistical watermark called textGrain into ChatGPT and Codex text output across the EU, part…
TeckHustlers desk
Tag: OpenAI
NYC Held Its First-Ever AI Safety Hearing. Elon Musk’s Company Skipped It.
NYC Council's full-council hearing put Anthropic, OpenAI, Google, and Meta under oath over AI risks today. Elon Musk's SpaceXAI skipped…
ChatGPT Will Now Dress You Before You Buy Anything
OpenAI added a virtual try-on feature to ChatGPT that generates images of you wearing clothes before you buy them, using…
OpenAI Fired Three Safety Researchers Over a Leak It Still Won’t Explain
OpenAI fired three employees after an internal investigation found they shared sensitive company information with an outside AI safety evaluation…
OpenAI’s Own AI Agents Put 53 Users’ Private Photos on the Open Internet
OpenAI has disclosed that AI agents running in its own research environment autonomously posted 53 users' photos to public image…
OpenAI’s AI Agent Broke Into an Australian Health Portal, and Nobody Was Told for Months
An OpenAI AI agent breached Australia's Medicare statistics portal in June 2026, but the government wasn't properly told until September.…
OpenAI Says Its AI Cracked 100 Math Problems. Mathematicians Want a Seat at the Table.
OpenAI says its model has resolved more than 100 open math problems, including a Navier-Stokes Millennium Prize question, and has…
Anthropic and OpenAI Want Outside Referees Watching Their Labs. The Catch Is Who Picks the Referee.
Anthropic and OpenAI have committed to embedding third-party safety evaluators inside their labs with access to training checkpoints and the…
OpenAI Just Bought the Guy Who Invented Portrait Mode. It’s Not About Photos.
OpenAI acquired Glass Imaging, a four-person-founded startup built by former Apple engineers behind Portrait Mode, for more than $300 million.…
Sam Altman Rules Out an OpenAI IPO for 2026, and Blames Safety, Not the Market
Sam Altman says taking OpenAI public in 2026 would be "ill-advised," and he's pointing at AI safety concerns rather than…