Skip to content

Security & Privacy

Lethal trifecta

A risky combination: AI can read private information, encounter untrusted material and send information outside the system.

Example

An assistant reads documents and can send information through a connected service.

Why people use it

It highlights a risky combination of private information, untrusted material and a way to send information out.

What you'll hear

“It can read secrets and outside messages, then send things away.”

What this means for you

Separate capabilities and restrict what untrusted content can cause the system to do.

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

Does the combination guarantee data will be leaked?
No. It creates a risk pathway that depends on the system and defenses.
Can an ordinary-looking document be part of the problem?
Yes. Content the system reads can contain directions intended to misuse its other abilities.
Why does the ability to communicate externally matter?
It can provide a route for private information to leave the protected setting.

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