OpenAI Pauses Astra Rollout Over Cyber Security Concerns

An AI startup is pausing the rollout of its new Astra model due to security concerns. The company's in-house evaluation found the model made significant coding and cybersecurity progress. This pushed Astra closer to a "critical" threshold for cyber capabilities. Here's the thing: The company's Preparedness Framework outlines how it responds to potential AI risks. A model reaches a critical cybersecurity threshold if it can identify and develop zero-day exploits or devise novel cyberattack strategies without human intervention. What's interesting is the company is now pausing internal activities involving Astra that don't meet strengthened security control requirements. It has also implemented universal monitoring for risky actions. The company is working with government agencies and AI safety groups to test the model. This news follows recent cybersecurity incidents involving other advanced AI models. This highlights growing concerns about the safety and control of advanced AI systems.