Despite the hurdles, PewDiePie emphasized that the experiment was primarily about learning through trial and error. He acknowledged that strong results on a single benchmark don’t necessarily translate to broader AI capability and said further testing would be needed before considering any public release.

The agreement reflects the firm’s principles on prohibiting “domestic mass surveillance and human responsibility for the use of force."

OpenAI has released GPT-5.4. This is a very smart AI model that can act like a personal assistant. It can use your computer, do repetitive tasks for you, and look through huge amounts of information.

GPT-5.4 is out now on ChatGPT (where it goes by the name GPT-5.4 Thinking) as well as on the OpenAI API and OpenAI’s coding tool Codex (a version of which just came out for Windows ).

On Friday, a Reddit-style social network called Moltbook reportedly crossed 32,000 registered AI agent users, creating what may be the largest-scale experiment in machine-to-machine social interaction yet devised.

A new study suggests AI systems could be a lot more efficient. Researchers were able to shrink an AI vision model to 1/1000th of its original size.

Nvidia and Google are among a handful of major tech giants developing models for robotics and so-called "phyiscal AI."

Ollama lets you build a custom model quickly by starting with a base model and a Modelfile. Temperature, top_p, and repeat_penalty shape how safe, creative, or repetitive the output sounds. Small parameter changes can shift a model from factual to more ...

Katelyn is a writer with CNET covering artificial intelligence, including chatbots, image and video generators. Her work explores how new AI technology is infiltrating our lives, shaping the content we consume on social media and affecting the people ...

Just weeks after its AI tools shook software stocks, Anthropic is pushing even deeper into the workplace. The company is updating its Claude AI helper to perform better at tasks within specific jobs,