EU proposes strict AI rules, regulating “high-risk” uses such as loan applications, with noncompliant companies to be fined up to 6% of annual global revenues
great message from Washington to Europe. The recognition of fundamental rights and dignity in trustworthy AI is pitch perfect. https://twitter.com/... Alessandro Mantelero / @mantelero : I completely agree. So far, several missed opportunities, with a text focused on technology assessment, rigid classification of high-risks, little consideration of safeguards for fundamental rights and freedoms. Another way is possible, based on HRIA, and we need to work on this. https://twitter.com/... Isaac Kohane / @zakkohane : Europe throws first serious regulatory punch for AI @nytimes https://www.nytimes.com/... Unclear if it finds the right tradeoff between enabling progress and mitigating risk but now ethical AI concerns are very much part of public policy. #ML #MachineLearning
Wall Street Journal
Context & Ripple Effects
This lands five days after Politico published the leaked draft proposing outright bans on social credit scores and surveillance uses with fines up to €20M or 4% of turnover. The formal proposal keeps the punitive architecture — now up to 6% of annual global revenue — and pivots from banned categories to a 'high-risk' classification covering everyday deployments like loan applications.
Reception splits along predictable lines: NYU's Isaac Kohane frames it as Europe throwing its first serious regulatory punch at AI, while Alessandro Mantelero argues the text repeats earlier missed opportunities — rigid technology assessment instead of rights-based impact assessment (HRIA). The gap between those two reads matters because it previews the compliance-clarity fight that follows.
First-order effects
Any company running AI in classified high-risk uses — credit scoring, hiring, essential services — must document and conform its systems or expose up to 6% of annual global revenue to fines; the compliance burden lands immediately on product and legal teams, not just on frontier labs.
US-headquartered AI providers are affected directly: the rules apply to systems deployed into EU markets regardless of where they are built.
Second-order effects
Vendors and consultancies gain a new market building conformity-assessment, audit, and documentation tooling around the high-risk classification — governance becomes a purchasable layer rather than an internal discipline.
Competing jurisdictions face pressure to answer with their own frameworks, since EU-market access becomes conditional on meeting European-defined risk categories — the classification itself becomes an export standard.
Third-order effects
The trajectory visible in this coverage runs from proposal (2021) through compliance guidance for systemic-risk models (2025 guidelines issued amid company complaints about clarity) to actual teeth: by August 2026, the EU can evaluate models before regional release and restrict market access outright — turning Brussels from rule-writer into pre-market gatekeeper.
If the pattern holds, 'high-risk' becomes the load-bearing category of global AI commerce: firms design products around which uses get classified, and regulators elsewhere copy the ex-ante evaluation model rather than post-hoc litigation.
The trend:AI regulation is shifting from drafting broad prohibitions to operating persistent, ex-ante market gatekeeping, with the EU moving first and other jurisdictions following its risk-classification template.
The United States welcomes the EU's new initiatives on artificial intelligence. We will work with our friends and allies to foster trustworthy AI that reflects our shared values and commitment to protecting the rights and dignity of all our citizens.
We want to use more #TrustworthyAI! For health, fighting climate change, convenience in everyday life -if we can trust #AI not to put our fundamental rights at risk. This is today's proposal - and that we become excellent in developing #TrustworthyAI https://ec.europa.eu/...
Today's proposal is a welcome first step towards regulation that we hope will protect people's rights while ensuring continued innovation and economic growth. We look forward to engaging with EU lawmakers as they shape the road ahead. #AI #TrustworthyAI #Facebook https://twitter.…
This morning, I read the new proposed @EU_Commission AI legislation, “Proposal for a Regulation on a European approach for Artificial Intelligence.” Here are my notes on the first 50 or so pages. https://digital-strategy.ec.europa.eu/ ...
The EU proposed #AI Rules. If passed, impact on #Microsoft #Google #Facebook! Failure to regulate would have had negative impact on EU strategic interests, commercial innovation already outpaces military R&D in many areas making AI a long-term principal for a geopolitical EU! htt…
Great, looking forward to seeing the details 👀 we need collaboration among democratic countries to governments AI and other technologies ↘️ https://twitter.com/...
Europe throws first serious regulatory punch for AI @nytimes https://www.nytimes.com/... Unclear if it finds the right tradeoff between enabling progress and mitigating risk but now ethical AI concerns are very much part of public policy. #ML #MachineLearning
Is this the first time that a US National Security Advisor has tweeted praise about an announced EU regulation package? I think so — great message from Washington to Europe. The recognition of fundamental rights and dignity in trustworthy AI is pitch perfect. https://twitter.com/…
I completely agree. So far, several missed opportunities, with a text focused on technology assessment, rigid classification of high-risks, little consideration of safeguards for fundamental rights and freedoms. Another way is possible, based on HRIA, and we need to work on this.…
Artificial Intelligence is opening new worlds for us. From treating diseases to better education and safer driving. The actions we present today will strengthen Europe's position as a global hub for excellence and trust in #AI. Stay tuned! #DigitalEU https://twitter.com/...
“It's going to make it prohibitively expensive or even technologically infeasible to build AI in Europe,” said Benjamin Mueller [of] the Center for Data Innovation. “The US and China are going to look on with amusement as the EU kneecaps its own startups.” https://www.wsj.com/...