A double-blind pilot puts closed-model evaluation inside a cryptographic boundary
DeepMind and external evaluation partners report a way to test a proprietary model without revealing either its weights or the evaluator's private prompts.
Newsroom / latest
Latest reporting
Evidence-linked signals and analysis about AI capabilities, work, skills systems, labour markets and governance — written for the decision that follows.
All dispatches
205 articles
EU AI Act Article 50: who must disclose what, to whom and when
The binding law separates provider duties from deployer duties and machine-readable marking from disclosures people can perceive. The Commission's guidance helps, but it is not the law itself.
When an AI capability test crossed into production: what the OpenAI–Hugging Face incident actually shows
Three counts describe three different parts of the incident. None supports the claim that 700 agents successfully hacked Hugging Face, but together they expose a wider evaluation-control boundary.
42% said they needed more AI skills; 15% had trained
Cedefop surveyed 5,342 employees in 11 European countries. The result describes self-reported need and training participation—not tested AI proficiency or one universal curriculum.
SAP documents a skills staging gate—and a trusted-source bypass
SAP's 1H 2026 documentation describes a review queue for imported and inferred skills, while trusted sources can write directly to the Attributes Library.
The AI Omnibus moved Chapter III Sections 1–3 duties for Annex III HR-AI to December 2027
Regulation (EU) 2026/1744 sets different dates for Chapter III, Sections 1–3 duties for Annex III and Annex I high-risk systems. Other AI Act clocks continue.
A Microsoft preprint estimates fewer small-group emails among selected heavy Copilot users
The study observes Microsoft 365 actions in 11 large companies. It does not measure completed work, productivity, collaboration quality or organisational transformation.