Speed ↓ · speed timing
Pause
Time is the binding constraint: alignment and governance can catch up if frontier training halts above some capability threshold.
Mechanism
Halt frontier training via activist pressure, open letters, or national moratoria until safety catches up.
What this name has meant
vintage driftThe name is stable; the content has shifted. A reader acting on the label without asking which vintage is being meant risks arguing with a position nobody currently holds.
Halt all frontier training via multilateral agreement. FLI open letter framing.
Most advocates now mean conditional pause triggered by eval thresholds, which is capability ceiling renamed.
If it succeeds: what binds next
Capability frozen at current levels. The binding question becomes when to unfreeze and who decides. A long pause decays into Abandon superintelligence by default.
A strategy that produces a worse next problem than the one it solved has not done durable work.
Falsification signal
Major states or frontier labs publicly defect from a declared pause, or verification tech cannot distinguish a real pause from a declared one.
A strategy held without a falsification signal is not strategy; it is affiliation. Continued support after this signal lands is identity, not bet. See the identity diagnostic.
Self-undermining threshold
overshoot riskWhen adopted by fewer than all significant actors.
Unilateral pause transfers leadership to defectors. The pausing actor loses influence over what comes next, the exact opposite of what the pause was supposed to buy.
Every strategy has a stable region where it reinforces itself and an unstable region where pursuit defeats it. The threshold between them is usually narrower than advocates acknowledge.
Historical analogue
Nuclear · Limited Test Ban Treaty moratoriumEvery strategy inherits a plausible ceiling from its precedent. The analogue conditions the realistic reach.
Produced
Atmospheric test halt among major powers.
Did not produce
Underground testing continued; holdouts (France, China) caught up.
People on the record
23Profiled figures appear first, with their tier in small caps. Each face links to the person and their full quote record. Tag: pause.
expertise mix · 20 profiled
recognition mix
A strategy whose endorsement skews to commentators or external-domain experts is in a different epistemic state from one endorsed mostly by frontier-builders. The mix is read carefully across both axes; see the board for criteria. Counts are over the 20 profiled people on this strategy (3 unprofiled excluded).
Anthony Aguirre
Expert in another field · Known across the AI/safety field
Aza Raskin
Governance, policy, strategy · Known across the AI/safety field
Connor Leahy
Deep ML / safety technical · Known across the AI/safety field
Daniel Kokotajlo
Governance, policy, strategy · Known across the AI/safety field
Eliezer Yudkowsky
Deep ML / safety technical · Mass-public recognition
Elon Musk
Public-square commentator · Mass-public recognition
Emad Mostaque
Public-square commentator · Known across the AI/safety field
Emmett Shear
Public-square commentator · Mass-public recognition
Geoffrey Hinton
Deep ML / safety technical · Mass-public recognition
Geoffrey Miller
Expert in another field · Known across the AI/safety field
Jaan Tallinn
Governance, policy, strategy · Known across the AI/safety field
Liron Shapira
Applied or adjacent technical · Recognised inside subfield
Liv Boeree
Public-square commentator · Known across the AI/safety field
Max Tegmark
Expert in another field · Mass-public recognition
Nate Soares
Deep ML / safety technical · Known across the AI/safety field
Rob Bensinger
Governance, policy, strategy · Recognised inside subfield
Steve Wozniak
Public-square commentator · Mass-public recognition
Tristan Harris
Governance, policy, strategy · Mass-public recognition
Yuval Noah Harari
Expert in another field · Mass-public recognition
Zvi Mowshowitz
Public-square commentator · Recognised inside subfield
Andrea Miotti
Founder of ControlAI; pause campaigner
Fynn Heide
AI safety engineer; PauseAI Europe
Holly Elmore
PauseAI US executive director
Load-bearing commitments
Worldview positions this strategy quietly assumes. If the claim fails empirically or philosophically, the strategy loses its target or its premise.
Time buys readiness.
Fails if: If alignment does not converge with more time, pause only postpones the decision.
Coordination on a halt is tractable at meaningful scale.
Fails if: If one actor defects, the halt burns its own advocates and clears the path for defection.
Coordinates
Conflicts, grouped by mechanism
2Lever opposition
same lever, opposite pullThe pair's primary lever is the same; they pull it in opposite directions. A portfolio containing both is internally incoherent on that lever.
Complements, grouped by mechanism
4Same phase, different layer
same stage, distinct leversBoth are active in the same phase of the transition but act on different layers (model vs institution vs culture). They cover different failure modes inside the same window.
Shared authority
same legitimacy sourceDifferent levers, same legitimacy source (democratic, state, technical, market). The pair hangs together under one kind of authority; it stands or falls with that authority.
Same-lever twins
4Both use the same lever in the same direction. Usually redundant inside a portfolio: each dollar or effort unit only buys one lever pull, even if two strategies are named.
Axis position
Source note: Pause as strategy.md