AGI Strategies
← overview

Time horizon ↑ · non preventive

Long reflection

Aligned superintelligence arrives before lock-in windows close and humanity can credibly commit to reflect rather than act.

Mechanism

Once aligned superintelligence exists, pause civilization's irreversible decisions for a long reflective period on values.

If it succeeds: what binds next

Reflection is happening. The reflector has to produce durable consensus; deep time value preservation paradoxes say the conclusions may be rejectable by the resulting population.

A strategy that produces a worse next problem than the one it solved has not done durable work.

Addresses 1 failure scenario

all scenarios →
Value lock-in by a narrow coalition

People on the record

3

Profiled figures appear first, with their tier in small caps. Each face links to the person and their full quote record. Tag: long-reflection.

expertise mix · 3 profiled

Builds frontier systems
0
Deep ML / safety technical
0
Applied or adjacent technical
0
Governance, policy, strategy
1
Expert in another field
2
Public-square commentator
0

recognition mix

Mass-public recognition
2
Known across the AI/safety field
0
Recognised inside subfield
1
Newer or less central voice
0

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 3 profiled people on this strategy (0 unprofiled excluded).

  • Anders Sandberg

    Anders Sandberg

    Expert in another field · Recognised inside subfield

  • Nick Bostrom

    Nick Bostrom

    Governance, policy, strategy · Mass-public recognition

  • Stewart Brand

    Stewart Brand

    Expert in another field · Mass-public recognition

Load-bearing commitments

Worldview positions this strategy quietly assumes. If the claim fails empirically or philosophically, the strategy loses its target or its premise.

Values

Reflection converges on better values over time.

Fails if: If reflection diverges or reveals irreducible disagreement, the strategy's premise fails.

Time

Indefinite delay is possible and productive.

Fails if: If delay cannot be coordinated or if reflection stagnates, the strategy degrades into abandon-superintelligence.

Coordinates

Primary leverTime horizon (Long view)
Acts onnon preventive
Coercionconsent
Actor in controlhumans
Time horizonpost transition
Legitimacy sourcedemocratic

Conflicts, grouped by mechanism

0

No strict conflicts catalogued. This strategy pulls a lever that nothing else pulls in the opposite direction.

Complements, grouped by mechanism

4

Stage-sequenced

one sets up the other

The pair is phase-offset: one acts before the transition, the other during or after. The first creates the conditions under which the second binds.

Alignment firstLegitimacy firstPlural AI ethic

Same-side diversification

same side, different lever

Both act on the same side (AI or world) but pull distinct levers. They cover several failure modes on that side while leaving the other side uncovered.

Irreducible human authority

Axis position

What the strategy acts onNon-preventive
Coercion levelConsent
Actor in controlHumans as principals
Time horizonPost-transition
Legitimacy sourceDemocratic

Source note: Long reflection strategy.md