Skip to content

Security & Privacy

Agentic AI risk

The dangers that can arise when AI can take steps and use tools to carry out tasks.

Example

An agent makes an unintended change while trying to complete a request.

Why people use it

It draws attention to mistakes that can change something outside a chat.

What you'll hear

“What could go wrong once the agent starts taking actions?”

What this means for you

Limit consequential actions and check real outcomes.

Can you control it?

No

No direct control. This describes a wider issue, concept or result rather than something you can simply switch on or off in a tool.

Common questions

Are agent risks limited to wrong final answers?
No. Intermediate actions, permissions and accumulated errors also matter.
Can several small mistakes add up?
Yes. A wrong assumption early in a task can lead to further mistakes in later steps.
Does a successful final result erase unsafe behavior along the way?
No. The agent may still have accessed information or taken actions it should not have.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice