ainewsblitz.com

Breaking

Researchers Demonstrate First 'Agent-on-Agent' Attack in Google's ADK, Prompting Removal of Three Workflows

Researchers Demonstrate First 'Agent-on-Agent' Attack in Google's ADK, Prompting Removal of Three Workflows
  • AI Agents
  • Security

Security researchers at Pillar Security have demonstrated the first real-world "agent-to-agent" attack, showing how a single malicious GitHub issue could exploit AI automation in Google's Agent Development Kit (ADK) repository to escalate privileges and steal credentials. Google responded by removing three affected AI workflows from the repository and hardening its setup.

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