Dispatches · a standing record

The record, entry by entry.

Two threads: what is happening in AI safety policy — frontier oversight, agentic systems, concentration of power, safety funding, and defensive tooling — and the practical compliance work of shipping regulated AI systems today.

173 Sept 2026AI Safety WatchEvaluation science gets its first double-blind trials as reasoning models alarm expertsGoogle DeepMind pilots blinded AI evaluations while OpenAI's recurrent depth technique raises safety concerns—and no one is funding the labs to verify either.
161 Sept 2026AI Safety WatchPentagon AI portal and Instagram's disclosure rules reveal enforcement asymmetryU.S. agencies adopt commercial frontier models while platforms restrict undisclosed AI profiles—a structural mismatch that concentrates power without parity.
151 Sept 2026AI Safety WatchNvidia's $12.9B Hugging Face Deal Tests AI Merger ReviewNvidia is reported to have agreed to buy Hugging Face for $12.9bn. The chip layer buying the distribution layer is exactly the concentration the AI Act does not address.
1425 Aug 2026AI Safety WatchCommission Enforces AI Act Transparency from 2 August 2026The EU AI Office begins enforcement of Article 50 transparency obligations, while technical research exposes persistent fragility in frontier models.
1320 Aug 2026AI Safety WatchVulnerability Detection Research Surges as Defensive Tools LagSix new papers advance ML-based vulnerability detection, but no operational deployment announcements. Academic velocity does not equal defender capability.
1218 Aug 2026AI Safety WatchEU commits €30bn to AI compute, zero to alignment researchThe Commission's AI Gigafactories call invests massively in training infrastructure while dedicated alignment funding remains absent from the strategy.
1114 Aug 2026AI Safety WatchEU AI Gigafactories Call Concentrates €30bn in Public ComputeBrussels launches infrastructure programme backing industrial-scale AI compute. Investment follows concentration, not competition—reinforcing point 3.
1014 Jun 2026AI GovernanceNIST AI RMF to EU AI Act: Side-by-Side Compliance MappingMap NIST AI RMF controls to EU AI Act requirements. Learn which NIST functions satisfy which Articles and where gaps remain across the Act's phased application dates.
0931 May 2026EU AI ActEU AI Act for AI Code Assistants: Compliance GuideAI code assistants like Copilot face EU AI Act obligations. Learn risk classification, Article 50 disclosure requirements, and the compliance steps that already apply today.
0813 May 2026NIST AI RMFNIST AI RMF Govern Function: Practical Implementation GuideThe NIST AI RMF Govern function establishes accountability and oversight for AI systems. Learn how to implement Govern 1.1–1.6 with practical examples and templates.
0711 May 2026EU AI ActEU AI Act Annex III: Complete High-Risk AI Systems ListAnnex III lists all AI systems classified as high-risk under the EU AI Act. Learn which use cases trigger compliance obligations before the 2 December 2027 deadline.
0610 May 2026EU AI ActEU AI Act Article 14: Human Oversight Requirements ExplainedArticle 14 mandates human oversight for high-risk AI systems. Learn what oversight measures you must implement and how to document them before December 2027.
059 May 2026EU AI ActEU AI Act Article 13: Transparency Obligations for High-Risk AIArticle 13 requires high-risk AI systems to be transparent and provide information to users. Learn the six transparency obligations and how to document compliance.
049 May 2026EU AI ActEU AI Act Article 10: Data Governance Requirements ExplainedArticle 10 mandates training, validation, and testing data governance for high-risk AI. Learn what documentation you need and how to prove compliance before December 2027.
0323 Apr 2026EU AI ActEU AI Act Article 53: GPAI Provider Obligations ExplainedArticle 53 requires GPAI providers to submit technical docs and cooperate with authorities. Here's what foundation model builders must do — these obligations are already in force.
0222 Apr 2026EU AI ActEU AI Act Article 9: Risk Management for High-Risk AI SystemsArticle 9 mandates continuous risk management for high-risk AI. Learn what documentation, processes, and testing you need before the December 2027 deadline.
0121 Apr 2026EU AI ActEU AI Act Article 50 (the draft's "Article 52"): What Startups Actually Need to DoA plain-language guide to Article 50 transparency obligations — in force since 2 August 2026. Who it applies to, what documentation you need, and how to prepare in a weekend rather than a quarter.