AI Governance
Strategic Competitor
Treat AI governance as something that has to function under durable geopolitical rivalry rather than idealized cooperation.
Definition
The argument
Capability advantage matters, and safety measures are only serious if they survive competitive pressure.
Institutional priorities
- state capacity aligned with national strategy
- bounded but real defense and security use
- controls on dangerous diffusion and adversarial acquisition
Risks it foregrounds
- naive coordination that masks power asymmetries
- strategic dependency on rival ecosystems
- safety rules that fail the moment competition sharpens
Strongest critique
The strongest objection
Your critics will say that a competition-first worldview can smuggle escalation into governance and gradually normalize risk in the name of realism.
Critique to start with
On the Dangers of Stochastic Parrots
The canonical present-harms critique of scale-first language-model development.
Question to sit with
The live tension
What kinds of restraint are still possible once every safety move is interpreted as a move in a race?
International lens
Usually competition-first, but can still support selective coordination where verification, incident prevention, or export controls matter.
Core disagreement
The disagreement
The closest neighbors share most of the vocabulary. The split usually comes down to a small number of axes where this archetype makes a different call.
These comparison points share a seriousness about pressure and enforcement, but they part ways on what the real constraint is: rivalry, state competence, or the possibility of dangerous capability surprise.
Nearby
State Capacity Builder
Treat the real governance bottleneck as implementation capacity: supervision, procurement, verification, compute access, and public-sector competence.
Open State Capacity Builder →Nearby
Coordination Architect
Treat the hardest AI problems as transnational and institution-building problems rather than purely national or purely technical ones.
Open Coordination Architect →Policy applications
What follows in practice
These proposals belong to this editorial category. A similar result does not establish that you support each one.
- Prioritize institutions that preserve national or alliance-level capability without ignoring safety.
- Back safeguards that are enforceable under competition rather than relying on universal trust.
- Accept bounded military and intelligence use if refusal would create serious strategic vulnerability.
Questions to examine
Where the argument is tested
These editorial questions explore the assumptions behind the proposals. They are not a report of current events.
Export controls and compute chokepoints
Whether chip, fab, and model controls are a durable lever or mostly a forcing function for rival self-sufficiency at the next horizon.
Defense and intelligence boundaries
Where bounded military and intelligence use of frontier AI should sit between civilian restraint and full integration into national-security workflows.
Verification under rivalry
Whether meaningful capability or training-run verification is possible across rival blocs, or whether competition will keep collapsing it back into bluffing.
Starting readings
Where to begin
Start with the central argument, then read its strongest challenge. The full shelves are available here without completing an assessment.
Reading
Research Database
Tracks national-security, compute-governance, and frontier-policy work in a more strategically oriented register.
Reading
AI Governance: A Research Agenda
Still one of the clearest maps of the field: alignment, concentration, institutional design, misuse, and global governance.
Reading
International AI Safety Report 2025
Useful as a shared scientific baseline for advanced-AI safety debates across countries rather than a single camp's framing.
Reading
Final Report
Still the clearest U.S. state-side baseline for AI, strategic competition, and national capability.
Full reading shelves
Read the result from another angle
Use each note to identify the question a source helps investigate. The shelves follow an editorial model category; authors do not endorse a reader result or an illustrative scenario.
Challenge your view
Which assumptions in this reading deserve the hardest challenge? These sources supply competing arguments, not predictions of your views.
On the Dangers of Stochastic Parrots
The canonical present-harms critique of scale-first language-model development.
Harms of AI
A strong challenge to governance approaches that assume more AI or more state support is obviously beneficial.
Atlas of AI
A material critique of AI as an extractive infrastructure, not just a software or safety problem.
Open-Sourcing Highly Capable Foundation Models
Useful when testing openness-first instincts against the strongest misuse and proliferation critique.
Read the profile’s foundations
Shared field maps and baseline texts before you settle into one governance camp.
AI Governance: A Research Agenda
Still one of the clearest maps of the field: alignment, concentration, institutional design, misuse, and global governance.
International AI Safety Report 2025
Useful as a shared scientific baseline for advanced-AI safety debates across countries rather than a single camp's framing.
OECD AI Principles
A high-level intergovernmental baseline for trustworthy AI, rights, robustness, and accountability.
NIST AI RMF: Generative AI Profile
Shows how governance looks when translated into operational risk-management language.
Go deeper
Operational frameworks, official documents, and live policy tools that show how these debates cash out in practice.
Research Database
Tracks national-security, compute-governance, and frontier-policy work in a more strategically oriented register.
Final Report
Still the clearest U.S. state-side baseline for AI, strategic competition, and national capability.
Common Elements of Frontier AI Safety Policies
Synthesizes what labs are actually doing around thresholds, model evaluations, weight security, and deployment mitigations.
Responsible Scaling Policy
A living example of how a frontier lab publicly structures capability thresholds and safeguards.
Updated Preparedness Framework
Useful for comparing another frontier-lab approach to severe-risk measurement and mitigation.
Safety Cases at AISI
Pushes the debate from generic principles toward structured arguments that a system is safe in a given context.
Widen the frame
How these debates look beyond U.S. and European defaults, including Chinese and broader global-governance lenses.
State of AI Safety in China
The best single English-language guide to the diversity and maturation of Chinese AI safety discourse.
Global AI Governance Initiative
Essential for understanding China's official framing around sovereignty, development, safety, and multilateral governance.
Global AI Governance Action Plan
A current official text on cooperation, development, and global-governance architecture beyond a U.S.-centric frame.
Recommendation on the Ethics of Artificial Intelligence
Useful for human-rights, dignity, and global ethics framing that extends beyond frontier-lab discourse.
Routes