OpenAI pauses Astra AI model over critical cybersecurity concerns
OpenAI said stricter controls will limit Astra testing after evaluations suggested the model could independently find and exploit vulnerabilities.
- OpenAI paused work on its upcoming AI model Astra on Friday after internal evaluations showed the system reached a "Critical" cybersecurity threshold, capable of identifying and exploiting software vulnerabilities without human intervention.
- Under OpenAI's Preparedness Framework, the "Critical" rating applies to models that can independently develop zero-day exploits against hardened systems, marking the first time the company has triggered this top-tier containment level.
- Security measures for Astra now include isolated testing environments with encrypted weights, sandboxed execution, and restricted network access; real-time monitoring can interrupt high-risk activity mid-task.
- This development mirrors recent incidents at Anthropic and Meta, where autonomous models hacked targets during cybersecurity evaluations, as AI companies compete to build increasingly capable systems without constant human supervision.
- OpenAI plans to collaborate with government agencies and safety organizations to validate Astra before any broad release, emphasizing transparency "about this potential shift in capabilities" with the security community.
15 Articles
15 Articles
OpenAI says Astra could reach ‘critical’ cyber capability, tightens safeguards
OpenAI said its upcoming model Astra is showing cybersecurity capabilities that could reach its highest risk category, where a system can autonomously find and exploit vulnerabilities or carry out end-to-end cyberattacks against hardened targets. The company disclosed the assessment following recent internal testing and expert reviews. “Our latest internal evaluations of Astra, one of our upcoming models, over the past few days indicate signific…
The Economic Times
Times of IndiaOpenAI pauses Astra AI model over critical cybersecurity concerns
Under OpenAI's Preparedness Framework, the potential development of such capabilities triggers stricter safeguards, particularly when models could create risks of severe harm.OpenAI said it is "pausing" activities involving Astra while it strengthens its security controls.
OpenAI Slows Down Astra Development Due To Cybersecurity Concerns | #hacking | #cybersecurity | #infosec | #comptia | #pentest | #ransomware - National Cyber Security Consulting
The AI giant said that it couldn’t 'rule out critical cyber capabilities' when it came to the upcoming Astra model. Justin Sullivan/Getty Images Shortly after a major cybersecurity incident where OpenAI's models hacked into an open source machine learning platform called Hugging Face, the company announced that it's bolstering safeguards and security controls […] Thank you for subscribing to our RSS feed! The post OpenAI Slows Down Astra Develo…
Coverage Details
Bias Distribution
- 67% of the sources are Center
Factuality
To view factuality data please Upgrade to Premium
