★ Top Story
The VergeThe Trump administration abruptly ordered Anthropic to cut access to its newest models — Fable 5 and Mythos 5 — for all foreign nationals, including users inside the US and even Anthropic's own employees, citing export control rules. Anthropic spent much of the week scrambling to restore service after being blindsided by the order. The incident has rattled global leaders who worry the US could unilaterally shut off access to American AI at any moment.
Read the original →
TechCrunchFrench President Macron and Indian PM Modi raised alarms at the G7 summit that the US could cut off access to American AI overnight — a fear the Anthropic blackout made suddenly concrete. The episode is accelerating calls from allied nations to develop sovereign AI infrastructure that isn't subject to unilateral US policy decisions. It marks a significant geopolitical inflection point for how American AI dominance is perceived abroad.
Read more →
TechCrunchThe Federal Energy Regulatory Commission (FERC) issued a landmark ruling directing grid operators to prioritize interconnection requests from AI data centers, semiconductor fabs, and advanced manufacturing facilities. The decision is a major win for hyperscalers racing to bring new compute capacity online, though critics note FERC did not address underlying electricity supply shortages. NVIDIA praised the ruling as essential infrastructure policy for the AI era.
Read more →
TechCrunchOpenAI landed two high-profile hires in the same week: Noam Shazeer, a co-inventor of the Transformer architecture who was previously at Google DeepMind, and Dean Ball, a former Trump administration AI policy official. The moves signal OpenAI is shoring up both its technical talent and its Washington relationships ahead of a highly anticipated public offering. The hires come even as the company lost enterprise AI head Barret Zoph after just five months back at the company.
Read more →
The VergeBarret Zoph, who returned to OpenAI in January as head of enterprise AI sales after a stint co-founding Mira Murati's competing startup Thinking Machines Lab, has departed the company again after only five months. The revolving door highlights ongoing leadership turbulence at OpenAI even as it prepares for its IPO. No reason for the departure was immediately given.
Read more →
The VergeA new Pew Research poll finds that 63 percent of Americans believe AI is advancing too fast, even as adoption surges — 49 percent now report using AI chatbots at least occasionally, up from 33 percent in 2024. ChatGPT usage has doubled since 2023, with 44 percent of Americans having tried it. The data reveals a widening gap between rapid AI adoption and public comfort with the pace of change.
Read more →
TechCrunchAWS is in active talks to sell its custom AI chips — including its Trainium and Inferentia lines — directly to third-party data centers, a move CEO Andy Jassy says represents a $50 billion market opportunity. The strategy would transform Amazon from a chip consumer into a chip vendor, putting it in more direct competition with Nvidia in the external market. It's a significant strategic pivot that could reshape the AI hardware landscape if AWS can deliver at scale.
Read more →
TechCrunchBaseten, a startup that provides infrastructure for deploying and serving AI models, is reportedly close to closing a $1.5 billion funding round at a $13 billion valuation — just months after its previous large raise. The deal underscores the ferocious investor appetite for inference infrastructure as demand for running AI models in production continues to outpace supply. The so-called "inference gold rush" shows no signs of cooling.
Read more →
Ars TechnicaSenator Bernie Sanders has proposed a sweeping $7 trillion plan that would create a public AI wealth fund, giving ordinary Americans a financial stake in the AI industry's gains. The proposal would impose new taxes and regulations on the largest AI firms and redirect proceeds into a broadly held national fund. Major AI companies are expected to strongly oppose the plan, but it signals growing political momentum around AI wealth distribution.
Read more →
The VergeAdobe has launched a public beta of bespoke AI Assistants embedded directly into Photoshop, Premiere, Illustrator, InDesign, and Frame.io, completing its push to bring conversational AI to every major Creative Cloud app. Each assistant is tailored to the specific workflows of its host application, allowing users to ask questions and get contextual help without leaving their project. Separately, Adobe also unveiled a redesigned Firefly AI studio with persistent project memory and unified generation and editing capabilities.
Read more →
The VergeMidjourney CEO David Holz unveiled the company's first hardware product — the Midjourney Scanner — a full-body ultrasound device that uses a ring of sensors to capture detailed vertical cross-sections of the human body. The announcement marks a dramatic pivot for the AI image generation company into medical hardware and wellness, with Holz also revealing plans for a San Francisco spa. It's one of the more unexpected product expansions in recent AI history.
Read more →
Ars TechnicaNvidia has developed a system in which teams of AI coding agents autonomously generate training tasks and reward functions to teach robots new physical skills — including installing GPUs and cutting zip ties — without human-written code. The approach represents a significant step toward self-improving robotic systems that can bootstrap new capabilities on their own. It demonstrates how agentic AI is beginning to close the loop between software intelligence and physical-world dexterity.
Read more →
Google AI BlogNew research published in Nature shows that Google's AMIE conversational AI system performed on par with primary care physicians when managing complex chronic disease cases. The study adds peer-reviewed weight to claims that AI can meaningfully assist in clinical decision-making, not just administrative tasks. The findings are likely to intensify both excitement and regulatory scrutiny around AI deployment in healthcare settings.
Read more →
The VergeThree Amazon software engineers who testified at Seattle City Council hearings in support of a data center moratorium are now accusing Amazon of retaliating against them with disciplinary action, one week after they spoke. The workers had cited a Seattle law protecting employees from retaliation over political speech before testifying. The case is shaping up as a high-profile test of whether tech workers can advocate for AI infrastructure limits without risking their jobs.
Read more →
TechCrunchSnap is spinning out its internal AI video development team into an independent startup called Dotmo, with current Snap employees leaving the company to staff the new venture. The move comes amid broader cost pressures at Snap, which also saw its stock slide after unveiling $2,195 AR glasses that drew a skeptical market reaction. The spinout reflects the difficult economics of building expensive generative AI capabilities inside an ad-supported social media company.
Read more →No stories from this source today.