75 articles
Google has entered a classified agreement allowing the US Department of Defense to deploy its AI models for any lawful government purpose, with non-binding limits on mass surveillance and autonomous weapons.
Brussels negotiators meet today aiming for a political deal on the AI Omnibus that would push the high-risk AI Act deadline back to December 2027 and lock in firm watermarking rules for synthetic content.
A U.S. State Department diplomatic cable instructs embassies worldwide to raise concerns with foreign governments about Chinese AI firms — DeepSeek, Moonshot AI, and MiniMax — allegedly extracting and distilling models from American labs.
ESMA Chair Verena Ross says the EU's markets watchdog is contacting supervised financial entities to test their defences as Anthropic's Mythos model raises the speed and reach of AI-driven cyberattacks.
The Justice Department moved to intervene in xAI's federal challenge to Colorado's algorithmic discrimination law on April 24 — the first time the DOJ has weighed in on a state AI regulation case ahead of the June 30 effective date.
China's National Development and Reform Commission has ordered top private tech firms — including Moonshot AI, StepFun and ByteDance — to refuse American capital in upcoming rounds unless explicitly cleared by Beijing, mirroring earlier US outbound-investment curbs.
Sheikh Mohammed bin Rashid unveiled a two-year framework to put autonomous AI agents in charge of half the UAE's federal government services, the most aggressive state-level agentic AI rollout to date.
President John Dramani Mahama presides over Ghana's National AI Strategy launch on April 24, unveiling a decade-long blueprint backed by a $250 million compute centre and continental ambitions across the AfCFTA.
A White House memo from OSTP director Michael Kratsios alleges China is running coordinated campaigns with tens of thousands of proxy accounts to distill US frontier AI models, landing three weeks before the Trump-Xi summit.
Federal disclosures filed for Q1 2026 show Anthropic spent $1.6 million and OpenAI $1 million lobbying Washington — both record quarters for the rival AI labs as Pentagon procurement, copyright and export controls dominate the agenda.
Anthropic is requiring government-issued photo IDs and live selfies from some Claude users to cut off access from US adversaries, according to The Information, as black-market relay platforms race to preserve Chinese developer demand.
YouTube is expanding its AI likeness detection tool to celebrities, talent agencies, and management companies, giving Hollywood a Content ID-style system for hunting down deepfakes of their clients.
German Chancellor Friedrich Merz called on the EU to exempt industrial AI from the bloc's regulatory framework at Hannover Messe, citing factory competitiveness ahead of the AI Act's August 2 application deadline.
An Axios report says the NSA has deployed Claude Mythos Preview for cybersecurity work even as the Pentagon, which oversees the agency, has formally branded Anthropic a 'supply chain risk.'
Nebraska's chief justice temporarily suspended Omaha lawyer Greg Lake after a brief he submitted contained 57 defective citations out of 63, including 20 fabricated cases generated by AI.
The European Commission has awarded its €180 million sovereign cloud tender to Post Telecom, StackIT, Scaleway and Proximus, closing a six-year procurement process intended to reduce institutional dependence on US hyperscalers.
Anthropic CEO Dario Amodei met White House Chief of Staff Susie Wiles and Treasury Secretary Scott Bessent on April 17, 2026, signaling a possible thaw in the company's Pentagon supply-chain-risk standoff.
The EU AI Act's August 2026 high-risk enforcement deadline for hiring tools and the Mobley v. Workday class action signal a new era of AI recruitment regulation.
After months of fierce debate, Linus Torvalds and kernel maintainers agreed on a policy that permits AI-assisted contributions like Copilot while forcing human submitters to take full legal and technical responsibility for any bugs, security flaws, or licensing issues.
Maine's LD 2082, which would prohibit the clinical use of AI in mental health therapy without a licensed professional, has been sent to Governor Janet Mills — part of a wave of state-level crackdowns on therapy chatbots.
Treasury Secretary Bessent and Fed Chair Powell convened an emergency meeting with top Wall Street banking CEOs to address cybersecurity risks posed by Anthropic's unreleased Claude Mythos AI model.
Florida Attorney General James Uthmeier opens a formal investigation into OpenAI, alleging ChatGPT played a role in the 2025 Florida State University mass shooting and poses ongoing risks to minors.
OpenAI testified in support of Illinois legislation that would limit AI developer liability even when models enable mass casualties or financial disasters exceeding $500 million.
Elon Musk's xAI filed a federal lawsuit challenging Colorado's AI anti-discrimination law, arguing it violates the First Amendment by forcing changes to how Grok generates responses.
OpenAI releases a 13-page policy document calling for robot taxes, a national wealth fund, and 32-hour workweeks as AI reshapes the economy.
Georgia's legislature adjourns with three AI bills on Governor Kemp's desk, including chatbot child safety requirements and a ban on AI-only insurance coverage decisions.
California Governor Gavin Newsom signed an executive order establishing AI procurement safeguards and granting the state authority to overrule federal supply chain risk designations targeting AI companies.
Apple briefly rolled out Apple Intelligence to Chinese iPhone users last week before pulling it back. The accidental launch — without regulatory approval — could expose Apple to administrative penalties under China's strict AI rules.
New Zealand startup ThroughLine, already a crisis contractor for OpenAI and Anthropic, is developing a hybrid AI-human system to redirect users showing violent extremist tendencies toward deradicalization support.
Apple removes the AI coding app 'Anything' for violating App Store rules, highlighting growing tension between fast-moving AI builders and mobile platform gatekeepers. The move raises questions about how AI-generated software will be treated on major app stores.
Innovation Council Action, a new political operation with the blessing of former AI czar David Sacks, is preparing to spend over $100 million in the 2026 midterms to elect candidates aligned with AI deregulation.
The Trump Administration released a comprehensive national AI legislative framework built around six objectives — from protecting children to winning the AI race against China — and explicitly preempts state-level AI laws it calls a patchwork threatening American innovation.
A federal judge has called the Pentagon's attempt to block Anthropic from government contracts 'classic illegal First Amendment retaliation,' marking a major legal victory for the AI company.
In his annual investor letter, BlackRock CEO Larry Fink warned that AI's economic gains will concentrate among a small number of companies and investors with data, infrastructure, and capital — potentially widening inequality rather than broadly sharing prosperity.
A federal judge in San Francisco has heard arguments in Anthropic's lawsuit against the DoD and will rule shortly on whether to block the company's military supply-chain risk designation.
Meta's Ray-Ban Display smart glasses face an unexpected hurdle in Europe: incoming EU regulations requiring removable batteries by 2027 may prevent the company from launching the AI-powered wearable in the region.
U.S. District Judge Rita Lin sharply questioned the Pentagon's decision to designate Anthropic a supply chain risk, suggesting the move looks like retaliation for the AI company's stance on autonomous weapons.
A federal court in San Francisco holds a pivotal hearing today in Anthropic's lawsuit against the Department of War, as the government argues the AI company poses an 'unacceptable risk to national security.'
Anthropic argued before a federal judge in San Francisco today for a preliminary injunction against the Pentagon's designation of the AI company as a supply chain risk to national security.
The Trump administration's seven-pillar AI policy framework moves to preempt state AI laws and direct the FTC to crack down on deceptive AI practices.
Michael Smith of North Carolina pleaded guilty to using AI to generate hundreds of thousands of songs and bots to stream them billions of times, fraudulently collecting over $8 million in royalties from Spotify, Apple Music, and Amazon.
Anthropic submitted sworn declarations from two senior executives contradicting the Pentagon's national security claims, revealing the two sides were 'very close' to agreement just one day after the supply-chain risk designation was finalized.
Federal prosecutors have charged Super Micro Computer co-founder Yih-Shyan Liaw and two associates with conspiring to illegally divert Nvidia AI servers to China, using elaborate schemes including dummy servers and removed serial numbers.
The Trump administration released a national AI legislative framework with six guiding principles, calling on Congress to preempt state AI regulations and pass uniform federal rules.
NHTSA has upgraded its Tesla Full Self-Driving probe to an Engineering Analysis after finding the system fails to detect degraded visibility from glare and fog, covering approximately 2.4 million vehicles.
Nearly 150 retired federal and state judges have filed an amicus brief supporting Anthropic's lawsuit against the Trump administration, calling the Pentagon's supply chain risk designation legally unprecedented and procedurally flawed.
The Trump administration filed a court brief defending its blacklisting of Anthropic as a supply chain risk, while former federal judges submitted filings supporting the AI company's legal challenge.
The Trump administration filed a 40-page court response arguing that Anthropic poses an unacceptable risk to military operations because the company could disable or alter its AI during wartime.
The US Senate Sergeant at Arms has cleared ChatGPT, Google Gemini, and Microsoft Copilot for official use by Senate staff, while notably excluding Claude and Grok from the approved list.
The Senate Sergeant at Arms has authorized three AI chatbots for official Senate work, notably excluding Claude and Grok from the approved list.
Washington became one of the first states to pass comprehensive AI transparency laws, requiring watermarks on AI-generated content and safety protocols for chatbots interacting with minors.
CENTCOM commander details how AI systems from Palantir and other contractors are accelerating target identification and decision-making in Operation Epic Fury.
Anthropic has filed lawsuits against the Department of Defense after being labeled a supply chain risk, escalating a dispute over military AI usage limits.
Caitlin Kalinowski, OpenAI's head of robotics, resigned over concerns about the company's Pentagon partnership, citing insufficient guardrails around surveillance and lethal autonomy.
The U.S. Supreme Court declined to hear Thaler v. Perlmutter, ending the legal battle over whether AI-generated artwork can receive copyright protection.
The US Department of the Treasury released a comprehensive AI risk management framework and shared lexicon to guide financial institutions deploying artificial intelligence systems.
The British government announces a new £40 million Fundamental AI Research Lab aimed at solving core AI limitations like hallucinations and unreliable reasoning while reducing dependence on US tech giants.
The Recording Academy announces that AI-generated music is eligible for Grammy Awards as long as a human author makes meaningful creative contributions, setting the first major industry standard for AI in music.
The FBI reports that AI voice cloning scams caused over $1 billion in losses in 2025, a 400% increase from the prior year, as deepfake audio tools become cheap, accurate, and widely available.
China's 15th Five-Year Plan places artificial intelligence at the center of its national strategy, calling for AI integration across all major industries and breakthroughs in quantum computing and humanoid robotics.
The European Union activates enforcement of its AI Act prohibition on social scoring, issuing formal warnings to three companies and signaling that the era of AI regulation with teeth has arrived.
Japan's Diet passes the AI Innovation Framework Act, explicitly permitting AI training on copyrighted material and creating a regulatory sandbox that has already attracted commitments from OpenAI, Anthropic, and Mistral.
Australia passes the AI Transparency Act mandating invisible watermarks on all AI-generated images, video, and audio. Platforms have 6 months to comply or face fines up to 5% of revenue.
California's AB-1892 requires all AI companies operating in the state to publish detailed summaries of their training data sources. The first US law mandating training data transparency takes effect January 1, 2027.
Over 300 Google employees and 60 OpenAI staffers urge their companies to draw the same red lines Anthropic did on military AI use.
Amazon, Google, Meta, Microsoft, OpenAI, and xAI sign the Rate Payer Protection Pledge, committing to self-powering AI data centers.
President Trump has ordered all federal agencies to stop using Anthropic technology after the company refused the Pentagon's demand to remove safety guardrails from Claude.
The EU AI Act's enforcement phase is now active, with the first investigations underway and penalties of up to 7% of global revenue. Here's what developers and companies need to know.
The Department of Defense fast-tracks xAI's Grok for classified military use while giving Anthropic 30 days to accept new contract terms or lose its defense partnerships.
OpenAI removed the word 'safety' from its mission statement during its for-profit restructuring. The old mission: 'ensure AGI benefits all of humanity safely.' The new version drops safety entirely.
The senator visited California to warn about AI's impact on jobs, wealth concentration, and energy consumption, proposing a freeze on new data center construction.
Both companies commit funding and participation to an international effort developing shared methods for testing and monitoring frontier AI systems.
Deadlines, risk categories, and penalties — everything companies deploying AI in Europe need to know about the EU AI Act. Non-compliance fines reach up to 7% of global revenue.
New regulations require all AI models serving Chinese users to pass a government security assessment, raising questions about open-source AI development.
The most significant federal AI regulation yet: mandatory safety testing, red-teaming requirements, and reporting obligations for frontier models. Here's what changes.