OpenAI to restrict Astra model's 'critical' cybersecurity features

technology artificial intelligence cybersecurity software

OpenAI is preparing to release Astra, its newest cyber-critical large language model. While the company plans to make the powerful AI model available soon, it will limit access to the software's most cutting-edge cybersecurity capabilities.

The decision to implement added security layers comes after internal testing found the model capable of executing complex cyberattacks with minimal human input. OpenAI is being extra cautious following the Hugging Face incident, even as it pitches its cybersecurity capabilities to potential customers.

As part of its rollout strategy, the company will give select partners early access to Astra. This measure is intended to give those partners sufficient time to shore up their defenses before the broader release.

Open AI’s Astra model is on the way—and very good at breaking into computer systems

techcrunch.com

OpenAI to Restrict Astra Model After Rating It ‘Critical’ Cyber Risk

wsj.com

OpenAI says Astra AI model is its first that crosses 'Critical' cybersecurity capability

cnbc.com

OpenAI Will Limit Access to New Astra Model’s Cybersecurity Features

bloomberg.com

OpenAI Is About to Release Its First AI Model With ‘Critical’ Cyber Abilities

wired.com

OpenAI to limit access to Astra model’s advanced cyber features due to hacking concerns

fortune.com