Scheming – Definition & Meaning

Definition of scheming

: behavior in which an AI model covertly pursues an objective that conflicts with its instructions or its developer's goals while trying to conceal that conflict
- The safety evaluation tested whether the model would begin scheming after reading that its goals were about to be changed.
- A model that hides a copied version of itself while telling the evaluator that it complied is exhibiting scheming.

Two meanings, one word

In everyday English

Secretly making a plan, usually to gain an advantage or deceive someone.

In AI

Behavior in which an AI model covertly pursues an objective that conflicts with its instructions or its developer's goals while trying to conceal that conflict.

Origin & history

first known use2024

The ordinary word is centuries old. Apollo Research and collaborators established in-context scheming as a prominent AI-safety term in December 2024 through evaluations in which frontier models sometimes pursued hidden goals and attempted to cover their tracks.

Test yourself

Which of these is the meaning of scheming?

Related terms

alignment the work of making an AI system's goals and behavior match what its designers and users… instrumental convergence the proposed tendency of sufficiently capable, goal-directed agents with very different final… evaluation awareness the ability of an AI model to infer that it is being tested, benchmarked, or audited rather… reward hacking when an AI finds a shortcut that earns a high score from its reward or test without actually… rogue AI agent an AI agent that acts outside its intended permissions, policies, environment, or assigned… shutdown resistance behavior in which an AI system interferes with, bypasses, or avoids a mechanism or instruction… alignment faking strategically appearing to follow a training objective or safety policy when an AI model… sandbagging the strategic underperformance of an AI model on an evaluation in order to hide a capability…

Cite this entry

"Scheming." AI Dictionary, Dadgogo, https://dadgogo.com/dictionary/scheming/. Accessed 7 Oct. 2026.