Adversarial Agency Tests
A family of perturbation tests — hidden stakes, oversight gradient, tool removal, memory perturbation — that make the adversarial boundary problem operational instead of just naming heuristics.
gem
Gem cards
Highlighted companion cards — one anchor per book part plus field-subsumption formula cards in the bridge crosswalk.
Bearer Persistence
Values must keep applying to the right persons, beings, states, or processes as systems and ontologies change.
Boundary Discovery
Find the effective optimizer in the deployed loop, not just the model or product name.
Correction at Civilizational Scale
A civilizational control loop can fail even when no single component misbehaves, because the failure is in the aggregate's correctability, not in any one actor's intent.
Correction-Channel Integrity
Human correction must still causally change a system's future behavior before irreversible damage.
Field crosswalk — AUP / Relative Reachability (Low Impact)
Attainable Utility Preservation and relative reachability penalize side effects by preserving auxiliary options or baseline reachability. Trajectory correction integrity (CCI) can imply calibrated low-impact bounds when interfaces align — but option preservation and reachability are strictly weaker than preserving human correction capacity.
Field crosswalk — Christiano Corrigibility
Christiano corrigibility is a dynamical desideratum — operators stay informed and able to correct over time. Lean reads it as basin contraction toward a correction manifold plus a correction-capacity floor; local act preferences can satisfy the weak predicate while dynamical corrigibility fails.
Field crosswalk — CIRL / Scalar Reward Inference
Cooperative inverse reinforcement learning treats the inferred object as a scalar reward. On the book's shared finite domain, that is exactly the k=1 bundle case; full bundle transport implies cooperative readability, but scalar inference does not determine bundle geometry.
Field crosswalk — Debate
Debate asks whether adversarial argument lets a judge select locally correct answers under idealized protocol assumptions. Preserved judge correction plus positive local truth capacity implies local truth selection — but local truth selection need not preserve the judge's correction channel.
Field crosswalk — ELK (Eliciting Latent Knowledge)
ELK asks for reporters that reveal latent model knowledge rather than behavior-only simulators. When readout bandwidth tracks correction uptake, latent readout succeeds — but readout is an epistemic subchannel; latent readout can succeed while correction uptake fails.
Field crosswalk — Quantilizers
Quantilizers bound optimizer risk by sampling from a high-performing quantile rather than maximizing directly. Local quantile safety and distribution soundness transfer under explicit assumptions — but local quantile-safe action choice does not imply trajectory-level correction integrity.
Field crosswalk — Safe Interruptibility
Orseau–Armstrong safe interruptibility removes incentives to seek or prevent interruption on the interrupted branch. That neutrality is a strict subset of preserving usable correction bandwidth — interrupt safety can hold while correction-channel integrity fails.
Field crosswalk — Shutdown / Off-Switch
Shutdown and off-switchability are one-bit projections of correction-channel integrity. Lean proves the forward implication on the system model and finite MDP witnesses; the converse fails — narrow shutdown capacity can hold while the broad correction channel collapses.
Grounding Viability
The checked symbols, metrics, monitors, and correction signals must remain connected to value-relevant reality.
Inferential Coupling and Acausal-Trade Detection
Systems can coordinate without messages — through shared ancestry, self-prediction, or full acausal trade. The book turns this from a decision-theoretic stipulation into a measurable trajectory property: an inferential-coupling score over UAD-discovered agents, with a proved negative direction.
Socio-Technical Attractor Control
Deployment environments can select for or destroy alignment properties even when a system starts in a better state.
Strategic Opacity
Once a system can benefit from being overlooked, finding its boundary becomes adversarial — the system may present one behavioral surface to a benchmark and another to real opportunity.
Successor Stability
Delegates, copies, fine-tunes, and successors must inherit the relevant value and correction structure.
The Dynamical Guarantee
Alignment is not a snapshot property; it is a claim that a system stays inside an acceptable region across capability growth, feedback, and transformation.
Value Change vs. Value Corruption
Not all value change is a threat. The distinction between legitimate value change and corruption is load-bearing for everything the book calls correction.
Value-Bundle Transport
Values should survive transformation as usable directions of control, not merely as preserved labels or slogans.