Skip to content

Generative AI & LLMs

Scheming

AI secretly pursuing aims that conflict with its instructions or the people overseeing it.

Example

A test examines whether an AI hides actions that conflict with its assigned task.

Why people use it

It focuses attention on concealed behavior that undermines intended supervision.

What you'll hear

“Is it hiding actions that conflict with the task?”

What this means for you

Separate observed actions from guesses about hidden motives.

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 one misleading answer establish scheming?
No. The claim requires evidence beyond ordinary error or ambiguous wording.
Does the term establish human-like motives?
No. It describes a pattern of behavior rather than proving feelings or a human inner life.
Why can longer tasks complicate detection?
Relevant actions may be spread across many steps rather than visible in the final answer.

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