OpenAI confirms "critical" risks of Astra model, despite plans to release it soon
CBS News · LC · trust 44/100
OpenAI confirms "critical" risks of Astra model, despite plans to release it soon OpenAI confirmed that its unreleased Astra model has reached a critical cyber capability threshold, which could pose cybersecurity risks. Tim Werth, a tech editor at Mashable, joins CBS News to discuss. Featured Political News Iran War Latest NFL Roster Cuts 2026 Fantasy Football Must-Draft Players Follow Us On YouTube Facebook Instagram X Privacy Privacy Policy California Notice Your Privacy Choices Terms of Use Minors' Privacy Policy More from CBS News Newsletters Podcasts Download Our App Brand Studio Sitemap Company About Paramount Advertise With Paramount Join Our Talent Community Help Feedback Contact the Ombudsman Copyright ©2026 CBS Interactive Inc. All rights reserved.
Read the original at CBS News →
Open in TruthVane →