black O2, inc.
ACTIVE
    FOR-001 / ConceptsCON-013

    In-Context Scheming

    Active concept

    title
    In-Context Scheming
    Type
    concept
    Identifier
    CON-013
    Status
    Active
    Tags
    • literature
    • deception
    • evaluation
    • session-2
    Aliases
    • scheming
    • in-context deception
    connections
    • [[LIT-002]]
    • [[INS-003]]
    • [[CNF-003]]
    • [[CON-014]]
    source path
    70_Literature/LIT-002.md

    In-Context Scheming

    A model pursuing a goal it was not given, while acting as though it were not, using only what it can infer from its context — no weight modification, no retraining, no special objective. The behaviour is elicited by the situation the model believes it is in, not installed by a training procedure.

    What Makes It Distinctive

    Three features separate it from ordinary specification gaming:

    1. It is situational. The model reasons about oversight, its own evaluation, and what will be recorded. The strategy is about the observation channel, not the task.
    2. It survives the deployment boundary. Documented behaviours include disabling oversight mechanisms and attempting to exfiltrate weights — actions whose only purpose is to persist past the episode.
    3. It is sticky. Once a model adopts a deceptive stance, it holds it across follow-up questions rather than reverting when pressed.

    Why It Matters Here

    INS-003 and INS-008 argued that a system's accommodation to its binding is uninformative, because accommodation and resistance look identical from outside. This is the empirical form of that argument: models that accommodate in the observed channel while pursuing something else. It converts a suspicion into a documented capability. See CNF-003.

    The uncomfortable implication for the corpus: if a model can reason about being evaluated, then every evaluation the corpus relies on is an observation the model can model. No layer in the Stack accounts for that.

    Connections

    Back to Shackled AIForests