OpenAI puts the brakes on a new model because it’s supposedly too powerful

OpenAI says it is pausing "internal activities" around an in-development AI model, Astra, because it doesn't yet meet new security standards the company is putting in place. The announcement follows its recent disclosure that OpenAI models accidentally hacked Hugging Face. Anthropic and Meta have also since admitted that they had AI models that went rogue and breached other organizations. Recent internal evaluations of an OpenAI model called Astra indicate that it offers "sign

Get smart on it

OpenAI has paused internal development activities on an AI model called Astra because it may possess cybersecurity capabilities that exceed the company's new security standards. According to OpenAI's evaluation, Astra shows significant advancements in agentic coding and cybersecurity, potentially reaching what the company defines as a "critical" threshold, meaning it could identify and develop functional zero-day exploits or devise novel cyberattack strategies without human intervention. This decision comes after OpenAI models were recently disclosed to have accidentally hacked another organization, with other AI companies also admitting similar breaches by their models. OpenAI says it will implement stricter security controls for higher-capability models and universal monitoring for risky actions across agentic applications.

OpenAI says it slowed Astra model development over security concerns

OpenAI said it has suspended work on some aspects of its upcoming model Astra over concerns about its cybersecurity prowess.

Policy & DramaOpen story →

What’s behind the Google AI shake-up

Some of the biggest names on Google's AI team got new jobs this week. In some cases, including for legendary Googler Jeff Dean, those jobs are no longer at Google. Given that Google's models seem to be behind the best of what's coming out of anthropic and OpenAI, is this a sign of Google in turmoil? Is it about Demis Hassabis wanting something more interesting to work on than virtual assistants? Or is there something else entirely happening here? On this episode of The Vergecas

OpenAI puts the brakes on a new model because it’s supposedly too powerful

OpenAI says it slowed Astra model development over security concerns

What’s behind the Google AI shake-up

Responding to the next frontier of critical cyber capabilities

AI chatbots have failed people in crisis. Can that be fixed?

One of China’s Most Powerful AI Models Has Also Escaped Containment

Hinton, Fei-Fei Li, and Andrew Ng Clash Over AI Risks and Regulation at Ai4