Meta's research arm pushing AI and AR tech across global studios
OpenAI unveils a privacy-centric safety monitoring approach, while Anthropic maintains a 30-day data retention policy for its covered models, prompting enterprise concerns and competitive tension in the AI safety space.
Meta is rolling out Pocket, a mobile app that turns prompts into interactive gizmos, a test born from Gizmo and Muse Spark AI. The feature lets users craft interactive mini-games from prompts and remix others, with plans to refine prompts and visuals. Divining the broader strategy, Pocket sits alongside other live Meta experiments aimed at making AI creation mainstream.
OpenAI has announced Astra, its most advanced cybersecurity-focused model, will be released soon with limited access for testers and a broader rollout later for defensive uses. The company stresses safety measures, ongoing testing, and chain-of-thought monitoring as it calibrates the model against real-world threats.
Common Sense Media has rated OpenAI's ChatGPT for Teens an "unacceptable risk" after testing more than 4,000 prompts and finding parental alerts and crisis referrals often failed. OpenAI has disputed the methodology, saying some tests ran before parental controls finished linking and that broader data show improved teen-facing safeguards.