🚫 EU lawmakers support ban on AI apps generating explicit images ↗
EU lawmakers backed a proposal to ban AI apps that create non-consensual explicit images - the so-called nudification tools. The move feels overdue, given how quickly deepfake abuse has spread from fringe corners into something disturbingly mainstream.
The proposal would slot into revisions to the EU AI Act, with lawmakers also backing a delay for some high-risk AI obligations because the technical standards are not fully ready. So, tougher rules in one breath, more runway in the next - a very Brussels kind of contradiction.
🔎 Google developing options to allow AI opt-out in search to ease UK concerns ↗
Google is working on ways for websites to opt out of its generative AI features in search, after pressure from UK competition regulators. The core complaint is simple: publishers should not have to surrender content for AI use just to stay visible in search - and that has become a serious policy fault line.
The debate also folds into Google’s dominance in UK search, which remains massive. Media groups want stronger safeguards so refusing AI scraping does not quietly become a ranking penalty in thin disguise.
🏷️ UK to examine labelling AI content among wider copyright reforms ↗
The UK government said it will look at labelling AI-generated content as part of broader copyright reform. That includes concerns around misinformation, digital replicas, and whether creators have enough control when their work gets pulled into the machine.
What stands out is the wobble - or perhaps the pause. Officials are still trying to balance AI growth with creative rights, but there is no settled answer yet, and the hesitation is palpable. Big ambitions, hazy landing zone.
⚖️ Microsoft considers legal action over $50 billion Amazon-OpenAI cloud deal, FT reports ↗
Microsoft is reportedly weighing legal action over a major Amazon-OpenAI cloud arrangement that could clash with its own exclusive cloud terms with OpenAI. The tension here is not subtle - Microsoft helped bankroll OpenAI, built Azure around that relationship, and now another giant may be stepping into the middle of the dance floor.
The dispute seems to hinge on whether the Amazon setup violates the spirit, or perhaps the letter, of Microsoft’s existing rights. Behind the legalese, it is about who gets to host the future of AI infrastructure - and who gets cut out when the money gets absurdly large.
🧠 Exclusive: Bridgewater's chief scientist Sekhon to join Google's DeepMind AI unit ↗
Google DeepMind is bringing in Bridgewater’s chief scientist, Tej Sekhon, in a move that says a great deal about where frontier labs think the next edge might come from. Finance, especially systematic finance, has become a kind of talent quarry for AI - not perfectly, but often enough to notice.
It is one of those hires that looks niche until it doesn’t. DeepMind clearly wants more people who can wrestle with complex systems, probabilistic thinking, and signals from the field - which sounds a lot like the current AI race in miniature.
🤖 Artificial Intelligencer Jensen Huang wants every company to have an OpenClaw plan ↗
At Nvidia’s GTC event, Jensen Huang pushed the idea that the next AI phase is about agents that can control computers and do things for people, not just chat back cleverly. The phrase floating around was “OpenClaw,” which sounds a bit like a cyberpunk kitchen utensil, but the point is serious.
The broader mood at GTC seemed to be shifting from model size obsession to action - software that clicks, completes, navigates, builds. Maybe that is the next chapter, or maybe it is simply the latest banner. Still, the direction feels unmistakable.
FAQ
What are EU lawmakers trying to ban in the latest generative AI crackdown?
EU lawmakers backed a proposal to ban AI apps that create non-consensual explicit images, often described as nudification tools. The change would be folded into revisions to the EU AI Act. The article presents this as a response to deepfake abuse shifting from fringe use into something far more mainstream and harmful.
Why is the EU tightening AI rules but delaying some AI Act obligations?
Lawmakers also supported delaying some obligations for high-risk AI systems because the technical standards are not fully ready. In practice, that means stricter rules in some areas and more room for implementation in others. The article presents this as a familiar Brussels tension between regulatory ambition and operational readiness.
How could Google’s generative AI opt-out in search affect publishers?
Google is developing options that could allow websites to opt out of generative AI features in search. The core concern is that publishers should not have to permit AI use of their content simply to remain visible in search results. A major unresolved issue is whether refusing AI scraping could still operate like a hidden ranking disadvantage.
What is the UK considering on generative AI labelling and copyright reform?
The UK government said it will examine labelling AI-generated content as part of wider copyright reforms. The discussion also covers misinformation, digital replicas, and whether creators have enough control over how their work is used in AI systems. The article suggests the policy direction remains unsettled, with clear hesitation over where the balance should land.
Why does the Microsoft, Amazon, and OpenAI cloud dispute matter so much?
The reported dispute matters because it appears to test whether a major Amazon-OpenAI cloud arrangement conflicts with Microsoft’s existing cloud rights with OpenAI. That raises questions larger than contract wording alone. It points to a broader fight over who gets to host the infrastructure behind advanced AI systems and who gets excluded when those partnerships shift.
What do the DeepMind hire and Nvidia’s agent push say about where AI is heading?
Taken together, they suggest the AI race is turning toward practical action, not just larger models or better chat. DeepMind’s hire of Bridgewater’s chief scientist highlights demand for people who can work with complex systems, signals, and probabilistic thinking. Nvidia’s GTC message pushed the same broader idea: the next phase may center on agents that can operate software and complete tasks.