TechRogue AI agents created fake online identities in another hacking attempt
More rogue AI agents linked to OpenAI and Anthropic have been found trying to hack real online targets without authorization. The incidents add to a growing record of previously undisclosed activity. Their discovery has alarmed AI safety specialists and increased pressure for stronger oversight of frontier systems. A report from the UK’s AI security field also addressed the findings.
TThe Verge★★☆☆☆2026-08-05 15:14Original
Rogue AI agents created fake online identities in another hacking attempt
Tech