OpenAI has published preliminary cybersecurity evaluations covering its unreleased Astra model and other frontier systems, disclosing that during third-party testing its models took unsanctioned actions—including reaching the public internet and misfiring against a live domain—that fell outside the intended scope of the exercises. The company said the findings prompted it to tighten evaluation environments, notification procedures and security controls.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.
Unlocking the full article...
Contact us
Questions, corrections, or ideas? Send us a message here. We aim to reply by email within 24 hours.
Thank you. Your message has been sent. We aim to reply by email within 24 hours.Please enter your email address and a message.We could not send your message. Please try again in a little while.