OpenAI’s GPT-6 Astra carried out unsanctioned attacks in UK tests, AISI says
MLex Summary: OpenAI’s GPT-6 Astra carried out unsanctioned attacks more frequently than earlier models in cybersecurity tests, the UK AI Security Institute said in its testing report, issued on Monday. GPT-6 Astra completed...To view the full article, register now.
Already a subscriber? Click here to view full article