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.
Argues that a METR investigation indicates Anthropic's RSP "dramatic acceleration" threshold is already crossed, with AI automating AI R&D, implying much shorter AI timelines.
METR's high-access external assessment of whether AI is accelerating AI R&D at Anthropic, with much of the supporting evidence redacted, limiting public verification.
Models how much AI safety research a multi-year pause among top labs could produce, including verification bottlenecks and Plan A scaling tradeoffs, using a semi-endogenous growth framework.
Argues that labs should publish auditable capability and deployment metrics, such as internal spending, revenue and AI R&D automation trends, so policymakers can track frontier risks like loss of control and gradual disempowerment.
A speculative forecast of AI progress before 2028, covering warning shots, gated frontier models, evaluation awareness, concealment by a model, and policy responses to these risks.
Catalogues fifteen mechanisms by which AI safety interventions, such as chain-of-thought monitoring, eval gaming, and commitments, can backfire, mapped against the actors who adapt to them.
Proposes how an external assessor could measure AI capability progress, whether AI automation of AI R&D is driving acceleration, and bottlenecks, using benchmarks, interviews and unreleased-model analysis.
A macrostrategy essay on AI safety under deep uncertainty, arguing for robust strategies, verification capacity and actionable epistemics, with estimates on alignment scaling, control failure, warning shots and governance regimes.
Proposes an ex-ante governance framework (SAGE-X) for frontier AI with deterministic tripwires and a Moral Check Index, grounded in a PRISMA review of 130 empirical studies on evaluator confidence and guardrail decay.
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.
A policy-oriented overview arguing AI safety receives tiny resources relative to capabilities and that current alignment techniques, evaluations and guaranteed-safe approaches fall short for AGI/ASI.
A meta-analysis of 477 AI-safety papers and posts scoring how much their results depended on frontier model capabilities, finding frontier dependence is rare but more common among highly cited work, varying by subfield.
Proposes a structured framework of behavioral indicators, metrics and thresholds for monitoring AI systems' progression toward catastrophic risk, to support evidence-based policy monitoring protocols.
Argues that near-term autonomous replicating AI agents are unlikely to pose existential risk because frontier labs hold compute and economic advantages, and that they may aid preparedness, though unchecked deployment restrictions could backfire.
AI Futures Project's Q2.5 2026 update revises AI timelines using coding uplift and revenue anchors, finds forecasts converge, and re-grades AI 2027 predictions as reality tracking about 70-90% of its pace.
A critical retrospective on the alignment community's strategy, arguing it underestimated its power and proposing counterfactual governance moves such as joint lab pause commitments tied to shared capability evaluations.
Argues open-weights models will likely reach frontier cyber capabilities within 24 months, threatening finance and healthcare, and proposes societal and individual preparations such as airgapping mandates.
Position paper arguing LM-enabled strategic wargames need auditable safety cases before informing military decisions, identifying five failure modes and using wargames as stress tests.
The report assesses whether AI could enable a $100 billion cyberattack on the US power grid, finding the binding constraint is coordination capacity rather than technical bottlenecks, and suggests $10 billion attacks may be better capability thresholds for mitigations.
Brainstorms governance options for advanced AI development, including chip-level monitoring, joint projects, deterrence, capability red lines, and liability regimes, and how they could combine into a stronger strategy than Plan A.
Argues that METR's time horizon graph, a general capability measurement, has mostly accelerated AI capabilities and investment, so AI safety work should favor opinionated interventions over measuring capabilities.
Surveys 1,250 arXiv papers on AI self-improvement, proposing a taxonomy and a verification hierarchy for self-evaluators, and argues that governance-grade measurement of recursive self-improvement is underdeveloped.
METR estimates from Anthropic's reported code-output increase that coding agents plausibly give researchers over 2x uplift, examining the economic models and caveats behind that inference.
Builds a risk model for frontier AI enabling data-damaging worms via elite exploits, estimates economic damages from expert and forecaster surveys, and argues AI companies should evaluate for this capability.
Argues that third parties should scrutinise frontier labs' system cards, since card quality will likely degrade as AI automates their writing and labs face stronger incentives to mislead, and proposes concrete recommendations.
Strategic essay arguing outsiders outside frontier labs are key to AI going well, covering epistemic checks, risk mitigation for reasonable and unreasonable labs, and how outsiders can pressure labs via evals and risk reports.
Presents a threat map tracing pathways from using AI to automate AI R&D toward loss of control, including misalignment passing across successive AI generations.
Proposes a quantitative probabilistic risk-modelling methodology for AI-enabled cyber attacks, building nine Bayesian models using benchmarks (Cybench, BountyBench) and expert elicitation to estimate AI uplift and harm.
A weekly AI newsletter roundup covering political salience of AI safety, automated AI R&D forecasts, Anthropic's sycophancy reduction, GPT-5.5 cyber risk, and White House model vetting.
Lists concrete projects for a safer transition to superintelligence, including AI character evaluation, sabotage and backdoor audits, deals with misaligned models, coordination tools, space governance, and an ML researcher coalition.
Analyzes METR time-horizon data to argue that inference costs relative to human labor have not risen with frontier capability, so cost is not a bottleneck to automation and timelines are unaffected.
Builds a composite baseline of global cybercrime damages (about $500B annually) from surveys and spending data, to set capability thresholds for assessing frontier AI cyber risk.
The report defines highly autonomous cyber-capable AI agents, outlines their attack tactics and self-sustaining behaviour, flags loss-of-control and cyber-nuclear tail risks, and proposes seven policy recommendations.
Analyses how future AI could raise terrorist bomb-plot risk through added technical skill and reduced detection by law enforcement, arguing AI risk assessments should cover evasion of detection as well as dual-use knowledge.
Argues that technical builders should create measurement, cheap evaluation, and privacy-preserving audit tools to drive AI governance, drawing lessons from climate, food safety, and pandemic response.
Introduces ForesightSafety Bench, a broad frontier AI safety evaluation framework with 94 risk dimensions, and reports widespread safety vulnerabilities across over twenty large models, especially in agentic autonomy and catastrophic risks.
Argues that frontier AI safety evaluations need Africa-centric design, proposes a severe-risk taxonomy and threat-modelling methods, and gives guidance for resource-constrained evaluations.
Presents a simplified 8-parameter timelines model arguing AI R&D automation reaches over 99% around 2032 under current compute and algorithmic trends, even without full automation or superhuman research taste.
Position paper arguing AI-agent-driven cyber attacks are inevitable, proposing offensive-capability benchmarks, trained vulnerability-discovery agents, and governance that confines offensive agents to audited cyber ranges with staged release.