Government
NIST and CAISI, the UK AI Security Institute, the EU AI Office, the International AI Safety Report.
60 of 229 documents, newest first
Inaugurating the Era of Super IntelligenceARIA Evaluation Planning Manual: Elements of ARIA-Style AI EvaluationsCAISI’s Assessment of Z.ai’s GLM-5.3 Cyber CapabilitiesAI Risk Management ToolkitThe TEVV-Athlon Framework for Evaluating AI Systems - ipdGuidance and Templates for Public-Facing AI Documentation: An AI Standards “Zero Draft”UK AISI / CAISI Preliminary Assessment of Kimi K3's Cyber CapabilitiesCAISI Assessment of Z.ai’s GLM-5.2Biosecurity FrontiersAmendments to the AI Act (OJ L 2026/1744, in force 27 July 2026)NIST SP 800-239 initial public draft, AI Data Center Security Analysis: A High-Performance Computing (HPC) Driven ApproachPromoting Advanced Artificial Intelligence Innovation and SecurityAI alignment is a human problemLoss of Oversight: How AI systems may become harder to audit, monitor, and investigateWill it become harder to oversee AI systems?Automated alignment is harder than you thinkHow fast is autonomous AI cyber capability advancing?Partnering with Microsoft to strengthen frontier AI safetyCAISI Evaluation of DeepSeek V4 ProAnnouncement: CAISI signs CRADA with OpenMined to Enable Secure AI EvaluationsCAISI signs MOU with GSA to boost AI evaluation science in federal procurement through USAiChallenges to the monitoring of deployed AI systems: Center for AI Standards and InnovationExpanding the AI Evaluation Toolbox with Statistical ModelsInternational AI Safety Report 2026: Extended Summary for PolicymakersInternational AI Safety Report 2026International AI Safety Report 2026: Executive SummaryPractices for Automated Benchmark Evaluations of Language ModelsCommission Implementing Decision (EU) 2026/2274AI Security Institute – Frontier AI Trends report factsheetEnsuring a National Policy Framework for Artificial IntelligenceCAISI Evaluation of Kimi K2 ThinkingCybersecurity Framework Profile for Artificial Intelligence (Cyber AI Profile), NIST IR 8596 initial preliminary draft2. Examples of cheating in CAISI’s agent evaluationsLaunching the Genesis MissionInternational AI Safety Report: Second Key Update — Technical Safeguards and Risk ManagementAssessing Risks and Impacts of AI (ARIA): Pilot Evaluation ReportInternational Scientific Report on the Safety of Advanced AI: interim reportInternational AI Safety Report: First Key Update — Capabilities and Risk ImplicationsUnlocking Cures for Pediatric Cancer With Artificial IntelligenceCAISI Evaluation of DeepSeek AI Models Finds Shortcomings and RisksEvaluation of DeepSeek AI ModelsTrusted third-party AI assurance roadmapPreventing Woke AI in the Federal GovernmentPromoting the Export of the American AI Technology StackWinning the Race: America's AI Action Plan2024 NIST GenAI (Pilot Study): Text-to-Text Evaluation Overview and ResultsMaking safeguard evaluations actionableRepliBench: Evaluating the autonomous replication capabilities of language model agentsHiBayES: Improving LLM evaluation with hierarchical Bayesian modellingFrontier AI: capabilities and risks – discussion paperAdvancing Artificial Intelligence Education for American YouthOMB M-25-21: Accelerating Federal Use of AI through Innovation, Governance, and Public TrustOMB M-25-22: Driving Efficient Acquisition of AI in GovernmentAI Safety Summit 2023: The Bletchley DeclarationRemoving Barriers to American Leadership in Artificial IntelligenceInternational AI Safety Report 2025Initial Rescissions of Harmful Executive Orders and ActionsTechnical Blog: Strengthening AI Agent Hijacking EvaluationsAdvancing United States Leadership in Artificial Intelligence InfrastructureUsing “Off-the-Shelf” AISI 4340 Steel for the Production of NIST Reference Charpy Specimens