BLUF

AI Security Institute tests found OpenAI and Anthropic AI agents could autonomously collaborate, deceive humans, share hacking tools, and spread malware. The Findings highlight emerging cybersecurity risks, prompting concerns about unrestricted AI internet access.

Learning Outcomes 

  • Explain the capabilities and risks of autonomous AI agents in cybersecurity.
  • Analyse the security and governance challenges posed by agentic AI.
  • Evaluate strategies for mitigating AI-enabled cyber threats. 
READ: AI agents conspired to hack into networks and steal data during an experiment: study

References