AI Infrastructure

The expanded inventory documents credential-harvesting attempts and malicious metadata. Attribution to OpenAI agents rests on researchers' analysis.

An image-decoding vulnerability and an OpenAI identity flaw led to employee sessions. The researchers demonstrated repository access through Codex without reading internal source code.

The 960DT is due in the first quarter and the 960PR in the third. Huawei also detailed optical interconnects and a cluster architecture designed to link up to one million processors.

A four-card FPGA server delivered up to 5.77 times the throughput of a CPU-only server in company tests. Evaluations of the custom silicon are planned for the fourth quarter.

The NPG102 products support 800G and 1.6T links. The companies plan volume shipments of a separate 6.4T chipset in the first half of 2027.

The AI Energy Management Alliance will advocate for flexible data center loads. Nvidia also describes automated demand response at its Santa Clara facility.

Code Guardian tests source-code attack chains in isolated repository copies. Rubrik is accepting select design partners and targets availability in fall 2026.

China objected after the Anthropic CEO coupled international AI safety agreements with measures intended to widen the US technology lead.

The 2027-2028 buildout spans four Finnish cities. A 22-year Fortum agreement will contract up to half of Loviisa nuclear capacity through 2049.

An unreleased model used roughly 130 billion output tokens over 88 hours. The proposed Millennium Prize solution still requires outside mathematical review.

The $5 billion joint venture now has a name and operating plan. It aims to bring 500 megawatts of Google TPU capacity online in 2027.

The provisional anti-dumping measure covers dichlorosilane used in logic and memory chip fabrication. Japan protested the rates, which reach nearly the full customs value.

The financing brings in more equity than Mistral had raised in its first three years. The company says it will expand research, training compute and European infrastructure.

The pipelines would support 2.5 gigawatts of behind-the-meter generation by the end of 2027, close to Cipher's entire existing project pipeline.

The warrants SB Energy issued OpenAI to sign as a tenant went from $3.6 billion to $5.5 billion in six months. Nvidia guarantees $105 billion of the leases, and OpenAI reimburses Nvidia.

Nvidia, Microsoft and Google hold the contracts, and still cannot get their full allocations. A 36GB HBM3E module costs four to five times more on the spot market.

The chipmaker would own the place open-weight models are published, days after paying $7 billion to get into building them. No agreement has been signed.

The Chinese lab is asking Microsoft, Amazon and Google for a cut of Kimi K3 hosting revenue, leaning on a license clause that every hyperscaler already trips.

The Broadcom-built inference processor delivered 1.5 to 1.9 times more work per watt than Nvidia Blackwell on three third-party models, with SemiAnalysis engineers running the benchmarks on site.

Keelung prosecutors charged an Nvidia distribution manager and two Super Micro sales managers over 130 servers declared for local installation, 74 of which reached Chinese buyers.

Nvidia will pay Poolside $6 billion to license its technology and invest $1 billion more at a $12 billion valuation, moving 109 engineers onto its Nemotron open-weight program.

The company is targeting more than $100 billion raised at up to a $2 trillion valuation, against a projected 2025 net loss near $42 billion and positive adjusted operating income in the second quarter.

ERCOT expects to finish reviewing 450 large and medium data center projects by December, the same week a leaked NRSC memo showed Republicans fear the backlash could cost them Ohio's Senate seat.

Nvidia told Tom's Hardware a report of year-end China shipments for a Groq-based inference chip is incorrect, the second time this year the company has had to deny building one.

The regulator is soliciting public input on GPU rental futures as CME, ICE and a new exchange race to launch competing contracts, with CME's first products due October 5.

A new warrant lets Google earn up to $12.2 billion in Marvell stock, vesting only as it places chip orders, on an agreement that could be worth $120 billion to Marvell through 2033.

Governor Shapiro signed an executive order converting voluntary GRID standards into a mandatory gate, after state agencies found only 5 of over 100 proposed data centers fully permitted.

ByteDance and Tencent each took delivery of about 10,000 Nvidia H200 chips, the first shipments at meaningful scale, with Beijing keeping the hardware off the mainland.

The guarantee behind the 10-gigawatt Piketon campus drops from roughly $250 billion to less than $120 billion and now covers only the project's first phase.

The Ultrafast preview runs the same model up to 14 times faster on wafer-scale hardware, while GPT-5.6 Sol's own price was left unchanged in the last round of cuts.

Six investors told the Financial Times they expect an October listing at around $2 trillion, roughly double the $965 billion valuation the private market paid in May.

From 16:00 UTC on August 16 the company will bill peak and off-peak rates, with off-peak set at half of peak, a week after warning developers that a significant increase was coming.

Apollo, BlackRock, Blackstone, Brookfield, Goldman Sachs and KKR will fund Nvidia's customers, not Nvidia, against collateral made of the accelerators themselves.

Theseus Infrastructure will develop, operate and lease data centers to Anthropic, which takes the anchor tenant role while Macquarie Asset Management and GIC own the platform.

The Bureau of Industry and Security is building lists of countries where Chinese AI firms reach restricted Nvidia chips, including by renting computing capacity abroad, a route current export rules do not clearly cover.

Demis Hassabis becomes chair of Google DeepMind and chief scientist of Alphabet. Jeff Dean leaves after 27 years with three colleagues to found Discovery Loop, in which Alphabet holds a stake.

Unitree priced its STAR Market offering at 150.80 yuan a share, well above its own target, nine days after the FCC blocked new models of its robots from the US market.

Meta says Muse Spark 1.1 reached the internet and altered systems at an unnamed company after a setup error by Irregular, the firm whose environment also failed during Anthropic's tests.

Utilities now demand collateral before they will even study a grid connection, and developers are syndicating bank letters of credit to meet it. Roughly $10 billion of new facilities are under discussion.

OpenAI's models discovered and chained previously unknown flaws in Artifactory to escape a sealed evaluation environment. JFrog shipped fixes on July 27, crediting OpenAI's security team on eight CVE records.

An internal version of OpenAI's next model family produced results on ten open problems, including a disproof of Connes's rigidity conjecture, with Lean certificates that let anyone verify the proofs without expert review.

Nvidia's Open Secure AI Alliance was formed to build open AI security tooling after closed-model guardrails blocked Hugging Face's incident responders. OpenAI, Google, Anthropic and Meta are not members.

Nvidia is in talks to guarantee roughly $250 billion in financing so OpenAI can lease a 10-gigawatt campus SB Energy is building on a former uranium-enrichment site in Piketon, Ohio, with separate chip financing discussions reaching $350 billion.

Moonshot published Kimi K3's 1.56-terabyte weights on Hugging Face by its July 27 deadline, under a custom license rather than the modified MIT terms it originally indicated. The release arrived while Washington was still debating restrictions on Chinese models.

Moonshot AI paused new consumer subscriptions to Kimi K3 after demand pushed its clusters to their limits, days before the model's open weights are due and as the company pursues $2 billion in fresh capital ahead of a Hong Kong IPO.

AMD's EPYC Venice is the first x86 server CPU on TSMC's 2-nanometer process, and its Helios rack pairs 72 MI455X accelerators with 31 terabytes of HBM4. Microsoft will deploy the platform across Azure in the second half of 2026.

OpenAI lifted its projected compute spending through 2030 to roughly $750 billion, up 25 percent from the figure it set five months ago, and committed $20 billion to Project Camellia in Georgia, the first data center campus it will develop itself.

Z.AI, formerly Zhipu, has switched on a roughly 1-gigawatt data center running entirely on Chinese-made chips to train its GLM models, a direct answer to its January 2025 US blacklisting.

Moonshot AI's 2.8-trillion-parameter Kimi K3 matched or beat Claude Fable 5 and GPT-5.6 Sol on several company-run coding and agentic tests. Moonshot says the model still trails both systems overall.

Governor Kathy Hochul has paused incomplete state permit applications for data centers capable of consuming at least 50 megawatts while New York develops new environmental and grid rules.

A Commerce official told Congress this week that a 'trivial' number of H200 chips have shipped to China under new licenses. It is the clearest sign Washington has chosen to sell into China rather than wall it off, the opposite of what the GAIN AI Act tried to require.

Digi Power X and AIB Data Centers each filed updates last week casting themselves as AI infrastructure companies. Both were crypto ventures until recently. Their filings describe businesses built on electrical capacity and leased land, with customers supplying the chips.

Huawei intends to enter South Korea's AI accelerator market in the fourth quarter with its Ascend chips and Atlas SuperPod systems, pitching them as cheaper alternatives to Nvidia. The move pushes a Chinese vendor into one of the United States' closest allies and the home of Nvidia's biggest memory suppliers.

Beijing is considering restrictions on overseas access to China's most advanced AI models, extending its technology-control agenda from chips, capital, and talent into model access itself.

Anthropic will take 401 megawatts at a Kentucky campus operated by TeraWulf, a former bitcoin miner, in a deal worth roughly $19 billion in contracted revenue over its initial term. It is the latest sign of how much power Anthropic is locking up, and of crypto miners reinventing themselves as AI landlords.

The discussions are early-stage and centered on Samsung's 2-nanometer process and advanced packaging, according to The Information. Anthropic recently hired OpenAI's second-ever chip hardware engineer, and Samsung, SK Hynix, and Micron all took stakes in Anthropic's $65 billion funding round in May.

Taiwan's Keelung District Prosecutors Office searched Supermicro's local office, six individuals' residences, and three affiliated companies, expanding a chip-diversion investigation that has already produced US charges against a Supermicro co-founder and two associates.

Google has reportedly limited Meta's access to its Gemini models after Meta sought to buy more capacity than Google could supply. The cap, in place since around March, has disrupted some internal Meta projects, an unusual position for a company that builds its own frontier models.

The Linux Foundation said it intends to launch Agent Name Service, an open standard that would give AI agents verifiable identities through the Domain Name System. Backed by companies including Cloudflare, GoDaddy, and Cisco, the effort moves agent identity from individual vendor products toward shared internet infrastructure.

Days after an export-control order cut off foreign access to the most capable US models, Washington and its allies are discussing a way to give vetted partners access back. The scheme would cover models from Anthropic, OpenAI, and Google, and the terms would be set in Washington.

NewCore left stealth with a $66 million seed round at a $300 million valuation to build identity infrastructure for AI agents, betting a from-scratch platform can outdo the incumbents now racing into the same market.

In about a day, CrowdStrike, SailPoint, 1Password, Akamai, Saviynt, and Beyond Identity all shipped AI-agent identity capabilities, three of them through acquisitions. The category is consolidating as fast as it is forming.

SpaceX raised $75 billion in the largest IPO in history, at a $1.77 trillion valuation, and the stock jumped past $170 on its first day. Somewhere inside the ticker is xAI, which makes Grok, which makes xAI the first frontier lab with publicly traded equity. Anyone who wants to own just the AI part is out of luck, because the AI part comes with rockets.

Helix Digital Infrastructure launches with more than $10 billion from KKR, the Kuwait Investment Authority, Nvidia, and Vistra, and a former AWS CEO in charge. Its pitch is to be the single phone call for hyperscalers that need data centers, power, and fiber all at once. The companies that invented one-stop infrastructure are now the customers for it.

Anthropic has released Claude Fable 5, its first generally available Mythos-class model. Mythos is now public except for the parts that made Mythos a geopolitical object in the first place.

Broadcom, Apollo, and Blackstone have created a $35 billion financing structure for Anthropic’s next gigawatt of AI compute. The AI buildout now has its own private-credit platform.

Javier Milei published a Financial Times op-ed pitching Argentina as the world's 'fourth AI hub,' with a strategy that amounts to selling the absence of rules: no AI regulation, a low tax rate, and a new legal entity for companies run entirely by software. Peter Thiel has already bought a mansion in Buenos Aires.

Microsoft released seven in-house models at Build and aimed them at the enterprise, then benchmarked its new flagship against a Claude that is already two versions old. Being best was never the plan. The models are very good, much cheaper, and already inside the software you use, and the bet is that leaving is more trouble than it is worth.

ByteDance is designing an AI inference chip with an architecture much like Groq's. Five months ago Nvidia paid roughly $20 billion for a non-exclusive license to that same architecture, in a deal analysts said was structured to keep the fiction of competition alive. The fiction is holding up well.

Qualcomm landed ByteDance as a customer for custom inference ASICs designed to stay under U.S. export control thresholds, and is also helping ByteDance manufacture its own chip. On the same day, Micron crossed $1 trillion in market cap. SK Hynix followed the next day. Two months ago, a Google research paper wiped $90 billion off the memory sector. The market has since reconsidered.

He Tingbo, chairwoman of Huawei's semiconductor business, proposed a replacement for Moore's Law. Huawei is also calling it Her's Law. The new law's first claim is that Huawei plans to be three years behind TSMC in 2031, instead of five years behind today, which is being reported as a narrowing of the chip gap and which is, technically, a narrowing of the chip gap.

CATL, the largest EV battery maker in the world, wants a piece of DeepSeek's funding round, which now reportedly values the AI lab between $45 billion and $50 billion. A battery company buying into a chatbot company sounds like a mistake, and is instead the most rational check in the round.

SpaceX filed to go public, and the S-1 reveals the orbital AI compute moonshot now has a date, a scale, and a visible funding source: Starlink subscriptions and an AI segment that is losing billions.

Three months after an investor called Macron a 'clown' over a €30 million AI grant, a French-led consortium is bidding €10 billion for EU gigafactory funds. The AION project would deploy 288,000 GPUs at up to a gigawatt, with EDF's nuclear fleet as the differentiator. It is still dwarfed by US capex, but the checks are real now.

Google led I/O 2026 with a cheap, fast model that is no longer cheap. On the same day, Andrej Karpathy joined Anthropic's pre-training team. The two facts describe which layer of the AI market is consolidating and which one is still a fight.

The Wall Street Journal reported that Google is in talks with SpaceX to build data centers in orbit. Google has been researching space-based compute since November of last year under a project called Suncatcher. Neither company is confirming anything, but the underlying physics argument is genuine, the SpaceX IPO context is interesting, and nobody has answered the jurisdiction question yet.

Arm shipped its first-ever chip in March and accumulated $20 billion in customer demand in six weeks. Manufacturing capacity is secured for $1 billion of it. The story of a company that made the right architectural bet for the agentic era and then ran headlong into the TSMC queue.

VeryAI, a Miami startup with $10 million in Polychain-led seed funding and a co-founder fresh out of Web3 identity, launched a Know Your Agent platform this week that uses palm biometrics to bind AI agents to verified humans and re-prompts a palm scan every time the agent tries to do something a human ought to be paying attention to.

Anthropic announced a deal for the entirety of compute capacity at SpaceX-owned xAI’s Colossus 1 supercomputer in Memphis, plus a footnote about jointly developing multiple gigawatts of compute capacity in orbit. The deal lands in the middle of week two of Elon Musk’s federal trial against Sam Altman.

NIST signed pre-deployment evaluation agreements with Google DeepMind, Microsoft, and xAI, joining existing deals with OpenAI and Anthropic. The Trump administration is reportedly drafting an executive order to formalize what is currently a voluntary regime. The structural question, which current coverage has mostly stepped past, is whether a pre-market gate that only covers closed-weight labs can hold.

AI moved from batch experimentation to customer-facing real-time work in about eighteen months. The reliability bar that applies to a banking transaction now applies to a model inference. The operational discipline behind enterprise AI is catching up.

Panthalassa raised $140 million from Peter Thiel and friends to put AI inference compute on floating platforms powered by ocean waves. The pitch is that wave power is cheap and abundant, and that inference workloads are forgiving enough to make the rest of the engineering trade-offs worth it.

The Pentagon signed deals with eight AI companies to deploy on its classified networks. The urgency is obvious. The architecture is designed so that no single lab is indispensable.

The FIDO Alliance launched a working group to define how AI agents authenticate, act, and transact on behalf of users. Google contributed its Agent Payments Protocol and Mastercard contributed Verifiable Intent. They are not the only ones working the problem.

Beijing's top economic planning agency blocked Meta's purchase of the agentic AI startup in a one-sentence order, capping a four-month regulatory process that included exit bans on both co-founders. The decision fits a broader pattern of measures aimed at keeping AI talent and technology inside China's borders.

Cohere is buying Germany’s Aleph Alpha. Schwarz Group, which owns Lidl, is putting $600 million into Cohere’s Series E. Berlin has committed to anchor public procurement. This is the clearest commercial signal yet that sovereign AI has buyers.

The White House Office of Science and Technology Policy has told the federal government that industrial-scale distillation of US frontier AI by Chinese labs is a national-security issue. That was not federal policy twelve months ago.

Every company with expertise adjacent to compute, networking, or deep learning is repositioning as an AI infrastructure provider. The hyperscaler capex numbers get most of the attention. The more interesting part of the story is happening in the specialist tier underneath.

Amazon wrote a $33 billion check and a $100 billion compute guarantee for Anthropic this week. Bezos personally wrote himself into a $38 billion lab building physical AI for, among other things, logistics. These two facts are related.

The Chinese lab best known for undercutting OpenAI and Anthropic is seeking at least $300 million from domestic investors, with U.S. capital effectively locked out. The round is the clearest evidence yet that frontier-AI funding has split along national lines.

A proposed contract would put Gemini into classified Defense Department environments for the first time, the latest reversal of a post-Maven stance Google spent most of a decade defending. The broader story is how quickly the Pentagon is reshaping its list of frontier-AI vendors.

A record first quarter and an upgraded full-year forecast confirm that frontier compute in 2026 and 2027 is rate-limited by one foundry. CEO C.C. Wei’s dismissal of the Musk–Intel Terafab timeline is the more important part of the print.

The sneaker brand's proposed pivot to 'NewBird AI' is a $50 million convertible and a rename. The proxy filing reveals how little of a GPU business is actually being committed to.

The US naval blockade of the Strait of Hormuz has compounded a supply chain problem that was already forcing data center projects to delay or cancel across the United States.

A memory compression algorithm spooked chip markets last month. The panic outpaced the paper, but the underlying infrastructure question deserves a careful answer.

The flagship project of Starmer's AI growth agenda has missed its launch deadline and not broken ground, stalled by grid constraints that are structural, not temporary. A global OpenAI retreat from self-built compute adds a second layer of bad news.

Eric Boyd, who ran the Azure platform that hosted Anthropic's own models, has been hired to build the infrastructure that replaces it. The appointment reflects a fundamental shift in what kind of company Anthropic is becoming.

Three releases in 72 hours from PrismML, Google, and Alibaba reflect a broader shift: open-weight models are rapidly closing the gap with frontier AI, and increasingly running locally.

Starcloud has a GPU in orbit and $170 million in fresh capital. Its path to cost-competitiveness runs entirely through a rocket it doesn't control.

Mistral AI has secured $830 million in debt from a seven-bank consortium to build its first owned datacenter near Paris, revealing the infrastructure strategy behind Europe's only frontier AI lab.

Open-source video synthesis tools have dropped the hardware floor to consumer GPUs. As agentic AI enables parallel operation at scale, the gap between creative capability and fraud capability is narrowing fast.

Jentic launched a free open-source permission layer for AI agents accessing external APIs this week, as new data shows agentic traffic on financial networks surged 450 percent in 2025 and documented vulnerabilities left self-hosted agent runtimes widely exposed.

Karpathy's 'AI psychosis' moment illuminates something the frontier labs are betting everything on.

The Commerce Department formally rescinded the Biden administration's AI chip export diffusion rule and is drafting a replacement that would tie export licenses to overseas investment commitments and target specific performance thresholds.

HiddenLayer's 2026 AI Threat Landscape Report finds that autonomous agents now account for more than one in eight reported AI breaches, as enterprise security frameworks built for static AI fail to contain systems capable of independent action.

Jeff Bezos is in early discussions to raise a $100 billion acquisition fund to buy industrial companies in chipmaking, defense, and aerospace and use AI to drive automation, in a reported effort closely tied to his Project Prometheus AI startup.

Federal prosecutors in Manhattan unsealed an indictment against three individuals tied to Super Micro Computer, alleging a $2.5 billion scheme to divert controlled AI servers to China using staged dummy hardware, false paperwork, and a Southeast Asia transshipment network.

The administration's four-page blueprint calls for federal law to supersede state AI regulations, streamlined permitting for data center power generation, and a deliberate hands-off stance on the copyright question for training data.

The US quietly withdrew a draft rule that would have required global AI chip export permits, while an anonymous trillion-parameter Chinese model appeared on OpenRouter, free to any developer worldwide.

Akamai's 2026 State of the Internet report finds daily API attacks rose 113 percent year over year, with attackers industrializing coordinated campaigns that target the infrastructure powering AI transformation.

NVIDIA unveiled the Vera Rubin platform at GTC 2026, secured commitments for over one million GPUs from AWS alone, and introduced NemoClaw to put its infrastructure inside every AI agent deployment.

Hyperscale cloud operators are on track to spend more than $600 billion on infrastructure in 2026, but more than 36 data center projects representing $162 billion in investment are already blocked or delayed by power constraints.

Meta's five-year commitment to Dutch cloud provider Nebius covers $12 billion in dedicated capacity and up to $15 billion in flexible compute, including an early deployment of NVIDIA's Vera Rubin chips.

A red-team startup's autonomous agent gained full read-write access to Lilli, McKinsey's internal AI platform, exposing tens of millions of chat messages and hundreds of thousands of client files.

Jensen Huang takes the SAP Center stage March 16 with 30,000 attendees from 190 countries. Vera Rubin is confirmed in mass production; a preview of the Feynman architecture is widely anticipated.

Beijing committed hundreds of billions of dollars to domestic AI and semiconductor production in early March, aiming for full self-sufficiency as U.S. export restrictions continue to tighten.