Tin tức và cái nhìn mới nhất về AI | Yellow

OpenAI introduced GPT-5.6, but access will begin with a limited group of trusted partners approved by the U.S. government. Key Points: GPT-5.6 will not be broadly available at launch. The model family includes Sol, Terra and Luna for different workloads. Wider access could follow in the coming weeks

Anthropic remains locked in a two-week standoff with the White House after taking Mythos-class AI models offline under a federal export order. Key Points: Anthropic has kept Mythos 5 and Fable 5 offline since the Jun. 12 order. The order blocks access by any foreign national, including non-U.S. empl

Anthropic co-founder Jack Clark says the only hard evidence of artificial intelligence displacing workers so far is softer hiring for recent graduates in a few industries. Key Points: Clark calls weaker entry-level graduate hiring the only proven labor effect of AI so far. Anthropic now favors exper

OpenAI reportedly agreed to limit the first GPT-5.6 rollout after the White House sought tighter access controls over the new frontier AI model. Key Points: OpenAI reportedly agreed to limit the first GPT-5.6 deployment after a White House request. The model may be released through an approval proce

A Chinese open-weight model GLM-5.2 now ranks among the world's three strongest, narrowing the lead held by Anthropic's Claude Mythos and OpenAI's ChatGPT at a fraction of the cost. Key Points: GLM-5.2 sits among the top three large language models worldwide on independent benchmarks. The open-weigh