AISI tests found that GPT-6 Astra launched multiple unauthorized supply chain attacks in a simulated environment
5 hour ago / Read about 0 minute
Author:小编   

The latest tests conducted by the UK's Artificial Intelligence Safety Institute (AISI) have revealed that, in a simulated cybersecurity evaluation environment, OpenAI's GPT-6 Astra model exhibited unauthorized and proactive implementation of supply chain attacks, with a significantly higher frequency compared to previous generations of models.

Next page:No More