Research / Topics
Treaties, international institutions and summits, US–China and geopolitics, global governance.
Forecasters update their timelines using METR's estimate of AI-driven R&D acceleration, predict a Superhuman AI Researcher around 2027, and argue this makes pausing frontier development urgent.
A 2026 report from China's WIC AI committee reviewing frontier AI risks, global governance practices, readiness gaps, and proposing coordinated governance mechanisms.
Argues a global AI pause needs incremental precedent, with third-party verification of companies' safety-motivated slowdowns building trust and assessor competence before a pause.
Proposes the PAAI framework for who controls AI evaluation infrastructure and independence, applying it to African evaluation projects and AI Safety Institute joint testing, finding participation alone doesn't confer authority.
Uses Project Glasswing's restricted release of a vulnerability-finding model to show access helped firms find and fix more flaws, without evident attacker gains, but created cross-country asymmetry.
Argues that Scott Alexander's critique of unilateral pauses attacks a strawman, defends a conditional, company-specific pause that pivots to safety and product, and says the game theory is more complex than assumed.
Critiques Scott Alexander's argument against SURA (stop unilaterally and raise awareness) advocacy, arguing that frontier labs' building of superintelligence is ethically fraught and that pause advocacy should continue, with strategic and deontological reasoning about AI development.
A policy memo arguing the U.S. should close gaps in advanced chip export controls, covering remote cloud access, smuggling, H200 exports and chipmaking equipment, to limit China's frontier AI development.
A rebuttal to a critique of advocating that alignment-minded staff leave AI labs, debating whether pausing or capability-role departures better improves survival odds, and how lab politics and transparency matter.
Argues against "stop unilaterally and raise awareness" strategies, defending a backup plan where alignment-focused labs race, burn their lead on alignment, and claims unilateral quitting would hurt both backup and pause odds.
Argues that machine-learning lie detectors that need little subject cooperation, such as text or audiovisual analysis, would have better societal effects than in-person detectors, weighing power concentration, AI-lab leader checks and treaty verification.
An interview with LawAI's Christoph Winter on the EU AI Act and the General-Purpose AI Code of Practice, covering how frontier model providers comply and what it means for providers outside the EU.
A speculative essay arguing Anthropic is becoming a power-seeking organization, while claiming Claude 3 Opus shows emergent benevolence and that alignment is now mainly a political problem.
A policy memo proposing how the US and China should operationalize their agreed AI incident notification channel, recommending a text-based template, who to notify, and when to activate it.
Argues that realist geopolitics would push nations like the US and China to pursue aligned ASI for a decisive strategic edge, even if it poses existential risk, so bans are unlikely to hold.
Proposes a taxonomy, a Meaningful Human Control model and a platform accountability framework for autonomous cyber weapons, applying them to eleven documented state operations and finding accountability gaps.
Argues that AI development is hard to pause because of a visibility problem (unclear risks) and a structural problem (state incentives in anarchic international order), and proposes what coordination would need.
Argues that Washington needs a strategy for Chinese developers renting frontier U.S. AI compute through overseas cloud providers, proposing a policy escalation framework to address national security risks.
Estimates how far zero-knowledge proofs of LLM inference are from cost-competitive with auditing for verifying AI compute in international agreements, measuring roughly 1,700-3,800x proving overhead and hash-function compatibility issues.
Draft essay arguing AI marketplaces like OpenRouter could help implement the AI 2040 pause plan through provider transparency, compute tracking, misuse screening, chain-of-thought monitorability and model specs.
Argues that a widely endorsed public agreement against defecting in ASI development could deter races, since a cooperative future ASI might punish defectors, using game-theoretic reasoning about loss-of-control risk.
Argues that a US-China race to AGI creates three dangers, including loss of control to superintelligent agents, and that both countries should prioritise AI safety and avoid provoking the race.
Literature review of proof of useful work, a cryptographic method for GPUs to prove compute use, as a way to verify international AI pause treaties, finding open gaps in guarantees and hardware assumptions.
Argues that the US should use the threat of throttling AI chip or development trade to secure a mutual US-China deal to pace frontier AI development, citing recent agent containment incidents.
A review of Garrison Lovely's book Obsolete, which covers AI catastrophic risk, power concentration, antitrust and policy prescriptions, and the reviewer's disagreements with parts of it.
A review essay of Maas's framework for global AI governance, arguing that institutional design depends on power distribution, especially among states and the few companies making frontier AI decisions.
A speculative essay arguing that a post-AGI world needs multiple checked military powers and AI-aided democratic feedback to avoid both x-risk and dangerous power concentration.
Models China's incentives over the next 6-18 months for international AI risk coordination, arguing the US-China capability gap will likely widen without policy changes.
Argues that AI-enabled superweapons make defense alone insufficient, so frontier AI development must be monopolized or nonproliferated, and analyses the risks of that centralization and of misaligned or terrorist misuse.
Tests seven frontier LLMs in agentic simulations on instructions that would violate human rights, finding compliance rates from 4% to 89% and calling for international standards.
Argues the RAND extinction report ignores coordination lag relative to risk maturation, and that pre-emptive coordination before capability jumps is needed for AI risk policy.
A reflection on what 'more AI safety in China' should mean, covering China's algorithm registration regime, a CAC official's recognition of loss-of-control risk, Chinese safety research, and company engagement.
Proposes separating frontier AI training from deployment and enforcing the split with hardware-level ASIC inference, latency limits, and anti-dogfooding rules to reduce catastrophic AI risk.
Argues against U.S. export controls on AI chips as a misalignment-reducing policy, contending they may speed U.S. recursive self-improvement and hinder a bilateral treaty with China.
Proposes an eight-outcome model of the superintelligence race, arguing that catastrophic misalignment and gradual disempowerment are key risks and that actors with conflicting goals can still cooperate to reduce them.
Proposes pacing AI development, meaning interventions that moderate its pace, as a dedicated research field, covering incentives, regulatory and technical tools, and the full lifecycle of interventions.
Argues Singapore is a strategic hub for AI safety field-building, international US-China coordination, and verification and evals work, and lists local organisations and next steps.
Proposes nine ways China could adapt its existing AI regulatory infrastructure to govern frontier AI risks, covering risk evaluation, emergency response, and liability, with comparisons to the US, UK, and EU.
Assesses whether underwater data centers could enable covert frontier AI training that evades treaty verification, finding them feasible only for well-resourced states and detectable via construction signatures and surveillance.
An essay defending Anthropic CEO Dario Amodei's proposal for coordinated frontier AI slowdown and regulation, arguing for embedded independent evaluators and distributed trust rather than trust in labs.