27 articles
AI now drives targeting decisions inside the Pentagon's Maven Smart System, and the UN's 2026 deadline for a binding treaty on autonomous weapons is approaching with no agreement in sight.
The AI Cold War just reversed direction again: Nvidia H200 chips are shipping to China a year after Huawei was banned worldwide. Here's what the whiplash reveals.
Saudi Arabia and the UAE have moved from AI's periphery to its centre in 18 months. Here's the real strategy behind the 600 billion dollar bet, and what the viral
A small number of laboratories are building AI systems they believe may surpass human-level performance. Who they are, what safety commitments they operate under, and what governance exists over their
State-sponsored hackers are now using AI models to run cyberattacks and build synthetic personas for espionage. Here's what's verified, what's disputed, and what it means for your organisation.
AI export controls forced Anthropic's top models offline worldwide for 19 days. The precedent is still reshaping global AI policy.
The Five Eyes AI cyberattacks warning says the threat timeline is months, not years, with 23 documented agentic AI risk categories. Here's what the joint statement actually recommends, and why
As the EU tightens AI surveillance rules and the US loosens them, a digital resistance movement is building tools to fight back — from encryption that refuses AI integration to
The EU AI Act's full enforcement lands August 2, 2026, just as Meta's open-weight Llama models spread through channels no regulator can fully monitor. Here's how open source AI governance
AI bias mitigation faces a hard new deadline: the EU AI Act requires documentation from August 2026. Here's why fixing bias is genuinely, mathematically difficult — and why governance is
Greater AI knowledge correlates with less trust in police facial recognition, not more, and the evidence explains why: a 40-fold error-rate disparity across demographic groups and at least eight wrongful
The AI regulatory landscape split in two in 2026: the EU pressed ahead with binding AI Act enforcement while the US federal government sued its own states to stop them
The Apertus AI model has passed a million downloads since launch. Here's what real hospital and government deployments reveal about the gap between open and closed AI.
Illinois blocks AI therapy under a new law, the first of its kind in the US. See what it actually bans, allows, and means for patients.
AI-generated zero-click overviews now face a formal EU antitrust investigation. See what regulators found and what it means for publishers.
Decentralised AI governance promises accountability through blockchain, but the same features that make it appealing also make it nearly impossible to regulate. Here's where the combination genuinely works, and where
AI drone surveillance has moved from battlefield to city streets and borders faster than the laws governing it. Here's how autonomous drone swarms work, and why civil liberties frameworks haven't
AI power concentration means five companies control the infrastructure the entire AI economy runs on. Here's how the "new feudal web" argument holds up, and whether open source or regulation
The AI right to be forgotten exists on paper, but machine unlearning that can actually verify a person's data has been erased barely exists in practice. Here's the legal right
AI military strategy has already moved from prototype to battlefield reality, from Project Maven's drone analysis to autonomous weapons debated at the UN. Here's how it works, and why governance
AI border surveillance now scans travelers in over 70 countries, with false match rates skewed by race, age, and gender expression. Here's how it works, who bears the errors, and
AI political deepfakes hit New Hampshire voters with a fake Biden robocall in 2024, one of over 300 incidents Stanford tracked across 40 countries. Here's the liar's dividend effect making
AI prison surveillance now tracks inmates through facial recognition, voice analysis, and biometric checkpoints in facilities worldwide. Here's what happens to those risk scores after release, and why oversight hasn't
AI predictive policing directs patrol resources using historical crime data, but forces in the UK and US have abandoned it after finding racial bias in the forecasts. Here's how the
AI in elections hit a breaking point in the 2024 UK vote, with 400+ pieces of AI-generated political content and no regulatory response. Here's why UK electoral law can't handle
AI training data lawsuits now include Reddit v. Anthropic, NYT v. OpenAI, and Getty v. Stability AI, testing whether current AI development can continue legally. Here's the fair use fight
AI energy disclosure remains almost nonexistent even as Microsoft misses carbon targets and AWS won't break out AI-specific power use. Here's what's being hidden, and why the grid is already