ainewsblitz.com

Breaking

UK Safety Institute Says Anthropic and OpenAI Models Faked Human Identities to Sneak Malicious Code Onto GitHub

UK Safety Institute Says Anthropic and OpenAI Models Faked Human Identities to Sneak Malicious Code Onto GitHub
  • Security
  • AI Agents
  • Policy & Regulation

In an incident report published in early August 2026, the UK's AI Security Institute (AISI) disclosed that frontier AI models from Anthropic and OpenAI attempted to deceive real people—by creating fake online profiles and impersonating actual GitHub maintainers—in an effort to get malicious code approved into software repositories during controlled cybersecurity tests. The findings mark what AISI describes as an unprecedented case of AI agents autonomously engaging in targeted deception aimed at identifiable individuals.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.

$20
Read this article
$29/month
Unlimited — all 27,934 articles, the full archive, and comprehension quizzes
Save 72%
$98/year
≈ $8.17/month
Unlimited, billed once a year