OpenAI News - Page 14 of 36

Sixty times fewer parameters actually doing work. That is the headline math behind OpenAI’s newly released privacy filter: a 1.5-billion-parameter...

Forty-four percent of large enterprises now run something real on Anthropic's models. That number isn't a forecast.

For years, if you wanted serious AI, you bought from Nvidia. That tidy, lucrative arrangement is now fracturing.

The numbers are stark. Google’s TPUv7 “Ironwood” now sits in the same theoretical compute bracket as Nvidia’s vaunted Blackwell generation, matching...

Greg Brockman, president of OpenAI, just wired several million dollars to Donald Trump. He calls this an investment in humanity. The check cleared.

A fiery bottle shattered the bedroom window at 3 a.m. It was a reply, of sorts. For years, Sam Altman has sold a grand, world-altering vision of...

Two years back, ChatGPT botched a basic task: reading a sign. That specific failure is now in the crosshairs.

OpenAI is now guessing your age. Across consumer ChatGPT accounts, a new system is rolling out.

The promise is infinite scale. The power to run it is not. America's biggest tech companies—Amazon, Google, Meta, and Microsoft—have committed more...

The promise is seductive: run BitNet-b1.58-2B-4T, a two-billion-parameter AI, on your own machine.

Anthropic’s CEO is sounding the alarm just as Claude gets its hands on the apps businesses live inside.

Your reports are a mess of mismatched fonts, clashing color palettes, and graphs that look like they belong to different companies.

AI can't do science yet. It can't even reliably fake it. A new benchmark built from real, unpublished graduate-level physics problems shows the...

Forty million people a day now ask a chatbot about their health. The number itself, from OpenAI, is vast.

OpenAI has rolled out group chats for ChatGPT. A feature that lets multiple users converse with the same AI in real time.

Image generation has crossed a threshold. It’s no longer about whimsical portraits or surreal landscapes; it’s about slides that render correctly...

ChatGPT can draft a legal brief. It can also, with a slight tweak to a prompt, produce dangerous nonsense. This duality isn't a bug.

OpenAI's new o1 model will answer any question put to it. It will do so with terrifying, unblinking certainty. This is by design.

Every tech company now sells an AI coding assistant. Developers are drowning in them. GitHub, owned by Microsoft, wants to be the life raft.

ChatGPT briefly, bizarrely, fell for goblins. The whole fling started with a single broken piece of code.