The National Cyber Security Centre has issued a blog post on managing the cyber risks associated with agentic AI. The post gives guidance on key considerations such as:
- Identifying what could go wrong
- Prompting carefully
- Setting the right level of oversight
- Controlling the AI agent’s environment with a robust sandbox
- Observability: logging, auditing and monitoring agentic AI activity as part of security operations
- Making your AI activity easy to attribute
- Emergency shutdown: maintaining the ability to ‘pull the plug’
This is a must read for anyone considering the use of agents!
You can read the full post here.
