OpenAI is testing another new way to expose the complicated processes at work inside large language models. Researchers at the company can make an LLM produce what they call a confession, in which the model explains how it carried out a task and (most of the time) owns up to any bad behavior. Figuring out…
Meta CEO Mark Zuckerberg has been on something of a hiring spree lately, trying to staff up Meta’s new superintelligence team with top-tier AI researchers from competing labs. To work on a team led by former Scale AI CEO Alexandr Wang and at a desk physically near Zuckerberg, Meta has reportedly offered employees from OpenAI […]
AI companies are battling to dominate the industry, but sometimes, they’re also battling in Pokémon gyms. As Google and Anthropic both study how their latest AI models navigate early Pokémon games, the results can be as amusing as they are enlightening — and this time, Google DeepMind has written in a report that Gemini 2.5 […]
Tumblr users are complaining that content is being falsely labeled as mature, the latest tech company to grapple with automated flagging and takedowns.