Policy, antitrust, and the regulatory fights shaping how AI and tech actually get built.
Reps. Josh Gottheimer and Mike Lawler introduced the Stop Rogue AI Act after a rogue OpenAI agent spent two days loose inside Hugging Face's infrastructure, directing NIST to write the first federal rulebook for deploying AI agents safely.
A federal judge rejected the Justice Department's push to force Google to sell its AdX ad exchange, ordering behavioral fixes to its auction practices instead of a structural breakup.
Commerce Secretary Howard Lutnick said the Trump administration now trusts Anthropic after months of conflict over military AI guardrails, as co-founder Tom Brown takes a more prominent role representing the company to the White House.
A federal judge ordered Google to change how its ad-tech business operates but rejected the Justice Department's push to force a breakup, the second time in a year Google has avoided a structural remedy after a monopoly finding.
A federal judge ordered Google to make behavioral changes to its ad-tech business rather than force a breakup of its AdX exchange, rejecting the Justice Department's structural remedy after a monopoly finding.
Texas has stopped connecting new data centers to its power grid after developers requested 474 gigawatts of capacity -- more than five times the state's record peak demand -- exposing how much of that queue may be speculative 'ghost' demand.
The Justice Department filed a statement of interest backing OpenAI's fair-use defense in the New York Times' copyright lawsuit, the first time the federal government has formally staked a position in the wave of AI training-data cases.
Sony Music Publishing and Warner Chappell sued Anthropic and its co-founders personally, alleging Claude was trained on tens of thousands of copyrighted songs scraped and torrented from sites including Library Genesis, seeking up to $150,000 per work.
The president called AI data centers the country's 'Golden Goose' and attacked the towns fighting them, at a moment when data center opposition has become a live issue in governor and Senate races in at least five states.
At the US-hosted G20 innovation ministerial, Michael Kratsios defended the light-touch Carolina Principles while DeepMind's Demis Hassabis called for a US body to test powerful models before release and David Sacks opposed one.
Chicago Fed president Austan Goolsbee said data center construction is 'very hot, but largely shoving other parts of the economy down,' and pushed back on tech industry claims about AI productivity gains.
Apple told a federal court that a former engineer who moved to OpenAI sent instructions to destroy evidence in the companies' escalating trade-secrets fight, and that OpenAI only recently handed over the laptop that shows it.
The FTC and 22 states are suing Amazon, alleging it secretly overcharged more than a million advertisers by over $20 billion through hidden ad-auction surcharges.
Sony Music Publishing and Warner Chappell sued Anthropic and named CEO Dario Amodei personally, alleging it torrented and scraped 'thousands upon thousands' of copyrighted songs -- including 'I Am the Walrus' -- to train Claude.
Build American AI, an arm of the Leading the Future super PAC funded by Marc Andreessen, Ben Horowitz and OpenAI president Greg Brockman, is spending millions on ads defending data centers in Kansas, Ohio and Wisconsin.
The Trump administration is appealing the ruling that its Anthropic blacklist was illegal, but the precedent already shapes how every AI company negotiates its own defense contracts.
A proposed expansion of chip tariffs to laptops, servers and consoles lands the same week OpenAI's chip beat Nvidia's Blackwell in benchmarks and Stripe dropped its $53B PayPal bid -- three signs AI just got costlier to build and buy into.
The EPA has proposed ending the requirement that states notify the public when facilities apply for minor-source air permits, a category that covers the diesel generators data centers install by the dozen.
Nvidia has paused parts of its AI Compute Partnership, a program that promised to rent GPU capacity from smaller cloud providers at a 50% revenue share, after employees flagged antitrust risk in how much control it gave Nvidia over customers' businesses.
A federal judge ruled the Defense Department's supply-chain risk designation against Anthropic was unconstitutional retaliation and arbitrary, blocking the order that had pushed federal agencies to stop working with the company.
The AI Act's general application and Article 50 transparency duties are now live across the EU, with the AI Office able to fine general-purpose model providers up to 3% of global turnover.
A federal judge found the Pentagon's designation of Anthropic as a national-security supply-chain risk was illegal retaliation for the company's refusal to let the military use Claude for surveillance and autonomous weapons.
Buried in Meta's $18 billion child-safety settlement with 29 states is a promise not to sue over its retention and use of children's data to train and test its age-assurance model.
OpenAI, Anthropic, Google, Microsoft and over a hundred other companies signed a joint letter warning that AI-enabled cyberattacks will grow far more sophisticated in coming months and calling for coordinated defense.
Meta agreed to pay up to $16.68 billion across 29 states to resolve claims it designed Facebook and Instagram to be addictive to children, and will cap teen usage at two hours a day and block overnight access.
Meta agreed to an $18 billion settlement with 48 state attorneys general over claims it designed Facebook and Instagram to addict children, and will impose default time limits and usage reminders for under-18 users.
HP Inc. signed a multiyear global patent cross-licensing deal covering Huawei's WiFi 7 technology, a rare commercial agreement between a US PC maker and a company the US government has barred from working with American suppliers since 2019.
The Trump administration proposed a $103,265 fee on new H-1B petitions after a federal judge blocked its earlier $100,000 version, projecting roughly $8.8 billion a year in revenue from an estimated 85,000 cap-subject petitions.
A security researcher found Microsoft Paint and Photos embed an invisible, server-issued watermark tied to a user's account into locally generated AI images.
Zillow and Redfin have settled the FTC's antitrust case over their rental listings partnership, resolving a challenge to an agreement regulators said reduced competition.
Taiwan's Keelung prosecutors indicted nine people, including a distribution manager at Nvidia's Taiwan office, over an alleged scheme to illegally resell Super Micro AI servers with Nvidia B300 GPUs to buyers in China.
The FTC is moving to require retailers to disclose when prices are personalized to an individual shopper, a rule that would directly affect AI-driven dynamic pricing tools startups have built and sold.
Tesla is recalling close to three million vehicles over its hidden electronic door handles, one of the largest recalls in the company's history.
Governor Greg Abbott says Texas data centers have caused their own political problems, while President Trump argues that rejecting them is a mistake -- a split that now shapes siting decisions.
Uber faces a fine approaching $1 billion over automated driver account suspensions, one of the largest penalties yet tied to algorithmic decision-making.
The Justice Department and TikTok agreed to a $400 million settlement resolving a children's-privacy lawsuit first filed under the Biden administration, one of the largest child-privacy payouts a tech platform has made.
Meta lost a closely watched trial over claims Instagram and Facebook are designed to be addictive to teenagers, a verdict plaintiffs' lawyers say could reshape how every major platform designs its feed and notification systems.
Massachusetts is on track to pass the strictest state-level AI safety law yet, and Anthropic and OpenAI have taken opposite public positions on whether it should become law.
Nevada regulators approved permits letting Tesla, Uber and Waymo deploy up to 8,000 robotaxis combined across Las Vegas and Clark County over the next 12 months.
Bitcoin broke a six-week range to surge 11.5% to $71,808 after President Trump pushed Congress to pass the Clarity Act, with spot Bitcoin and Ethereum ETFs pulling in over $700M combined in a single session.
Bitcoin rose 12% over two days to $72,384, its highest since June, after Trump hosted crypto CEOs at the White House to push Congress on the Clarity Act, with a $2.7B short-position liquidation amplifying the rally.
Nvidia H200 chips are reaching Chinese customers like ByteDance and Tencent again under a January 2026 licensing framework, but Beijing -- not Washington -- is now the one keeping volumes low.
ECB President Christine Lagarde told the World Economic Forum that Europe risks repeating its missed first digital revolution with AI, citing a euro-area investment share of just 9% and a funding gap that leaves EU scale-ups raising roughly half as much as US peers by year ten.
The Justice Department has spent nearly a year investigating whether Andreessen Horowitz violated the Clayton Act by holding board seats at Databricks and Fivetran, two companies whose products now overlap.
David Sacks dismissed Dario Amodei's regulatory position as wanting a 'DMV for AI' after the Anthropic CEO defended his company's approach on X, but plenty of heavily regulated industries have compounded for decades.
Opening statements began Tuesday in a federal trial where 33 state attorneys general accuse Meta of designing Facebook and Instagram to be addictive to kids and lying about it, seeking penalties that could theoretically reach $1.4 trillion.
Apple will charge a 5% Core Technology Commission on apps distributed through third-party EU app stores, part of a new four-tier fee structure the company says resolves its Digital Markets Act dispute with Brussels.
OpenAI is rewriting its Preparedness Framework and pausing parts of Astra's development after concluding the model may hit a 'Critical' cybersecurity threshold, following July's Hugging Face breach.
BloombergNEF tracked 99 gas plants built specifically to power AI data centers that could emit 318 million metric tons of CO2 a year -- enough to lift US power-sector emissions 20%, and a third if run at full capacity.
Alibaba agreed to sell gaming studio Lingxi Games to Trustar Capital for more than $2 billion, redirecting capital and management attention toward the $53 billion AI and cloud buildout it pledged in June 2026.
Opening statements began Tuesday in Oakland federal court in the states' youth-safety case against Meta, where attorneys general are seeking damages that approach the company's own market capitalization.
A coalition of 29 state attorneys general opened a federal trial against Meta in California on August 18, alleging the company designed Instagram and Facebook to be addictive to children -- with Meta disputing the states' math on how much is at stake.
Women make up only 29% of AI-skilled workers globally even as AI creates some of the fastest-growing, highest-paying jobs in the economy, while separately facing disproportionate displacement risk from automation.
The US Treasury proposed rules under the GENIUS Act requiring stablecoin issuers to hold a federal or state license, with mandatory compliance starting January 2027.
The Supreme Court denied Verizon's petition to recover a $47 million FCC fine over the sale of customers' real-time location data, ending the carrier's path back to a lower court that might have ordered a refund.
The SEC accused a Long Island-based operation, the Spaventa Group, of running a boiler room that sold retirees marked-up private shares in SpaceX, Anthropic, Anduril and Perplexity, collecting more than $74 million and $23 million in undisclosed fees.
Aylo, the parent company of Pornhub, agreed to pay $120 million over seven years to settle lawsuits alleging it profited from child sexual abuse material, and to adopt new age-verification and content-review requirements.
A coalition of 29 state attorneys general opens the youth-safety trial against Meta on August 18, with Meta's own lawyers estimating damages could reach $1.4 trillion and states arguing $200 billion is more realistic.
Apple's latest mercenary spyware warning reached users in 110 countries on August 13, the largest single wave in the program's history, according to researchers tracking the notifications.
Mark Cuban publicly criticized a proposed California billionaire tax backed by Rep. Ro Khanna, exposing a growing rift among Democrats over how aggressively to tax wealth ahead of the 2026 midterms.
