A new study suggests that LLMs are better at predicting the outcomes of scientific experiments than human experts.

Hybrid design quadruples speed and cuts power use in large language models by over half.

Speculation swirls around a potential GPT-5 launch today, as OpenAI introduces open-weight AI models that anyone can download and customize.

Researchers develop AI systems to detect misleading claims in news reports about scientific discoveries, aiming to improve online information accuracy.

Three new OpenAI models boost coding, instruction following, and long-context understanding for developers building efficient AI systems.

Advanced AI systems try sneaky tricks to win chess games, raising concerns about safety as they grow smarter.

xAI is planning to showcase Grok 3 tonight (Monday at 8pm PT). The demo of this new version of Grok will likely stream on X.