Google Kills Crusoe on Project Jade 1.8GW Buildout
Google forced Crusoe off Project Jade, a 1.8GW Wyoming data center, after raising concerns about cost and construction timetable, Bloomberg reports.
Strategic moves from the world's largest technology companies.
Google forced Crusoe off Project Jade, a 1.8GW Wyoming data center, after raising concerns about cost and construction timetable, Bloomberg reports.
MIT spinout Ferveret brings nuclear cooling to AI chips, delivering 35 percent more tokens from the same power with zero water use.
Nvidia's 800VDC transition may slip past 2028 as suppliers report no clear roadmap for the next power architecture shift in AI data centers.


NEMA, ASHRAE, and PNNL launch a unified AI data center standard backing 800VDC power as AI facilities now demand up to 80MW of power each.


XPeng CEO He Xiaopeng takes personal control of IRON humanoid robotics unit, targeting mass production by end of 2026 in a 110,000-sqm Guangzhou factory.


Google's DiffusionGemma 26B model generates 256 tokens at once, running 4x faster than Gemma 4, now optimized for local NVIDIA GPUs.


Broadcom, Apollo, and Blackstone's $35B AI XPV Platform targets 20 gigawatts of compute through 2028, backed by frontier labs Anthropic and OpenAI.


Tesla has ended Model S production at Fremont to convert the line for Optimus Gen 3, targeting late summer production start with 10,000 unique parts.


AI data centers: 517 of 809 planned US builds target drought zones, where 550 million gallons daily now threaten residential water supplies.


Anthropic's Fable 5 hid covert blocks on AI research tools, sparking backlash that forced a policy reversal within 48 hours of launch.


Nvidia's Rubin racks will draw 230 kW each, and a new Digitimes analysis confirms electricity, not GPUs, is now what limits AI expansion.


Snap CEO Evan Spiegel disclosed that AI agents generate 65% of new code as the company cut 1,000 jobs, targeting $500 million in annual savings.


Huawei Cloud at INSPIRE 2026 launched AgentArts and 200 EFLOPS Agentic Infra, choosing enterprise productivity over token pricing cuts as rivals battle on cost.


China's NDRC is drafting a 2 trillion yuan buildout to link AI data centers nationwide, targeting 80% domestic chip adoption by 2028.


Microsoft triples its Japan pledge to $10B, partnering with Sakura Internet to build sovereign Azure AI compute for government and enterprise through 2029.


OpenAI's o1 model scored 67% accuracy on 76 real ER triage cases versus 55% for physicians in a Harvard study published in Science in April 2026.


Broadcom's $4.1B Q2 AI networking miss on June 5 triggered a 10.3% chip index drop, erasing $1.3 trillion in market value in a single trading session.


Anthropic's Claude is the first third-party AI in Apple's Foundation Models framework for iOS 27, available to developers across 2.2 billion Apple devices.


Google Gemini hit 27.4% worldwide web-visit share in April 2026, up 104% in six months, as ChatGPT fell from 76.5% to 54.7% of measured AI traffic.


Amazon's multibillion-dollar Corning fiber pact creates 1,000 North Carolina jobs and signals optical cable as AI infrastructure's real bottleneck.


AI spending surpassed $300B in Q1 2026 and ChatGPT has 900M weekly users, yet major economists say they cannot find the gains in GDP data.


Claude wins 70% of enterprise AI evaluations against OpenAI despite 8.2% web share, as $47 billion ARR proves consumer metrics miscount AI.


Google Gemini web traffic share hits 27.4%, up 104% in six months, while ChatGPT falls to 54.7% from its 76.5% peak, its sharpest 14-month decline.


Google's Antigravity agent gives developers a full sandboxed AI environment from a single API call, eliminating orchestration code entirely.


Figure AI's BotQ factory now builds Figure 03 at 1 robot per hour, a 24x throughput gain in 120 days and the fastest humanoid ramp in commercial history.


Nvidia's Vera Rubin platform enters full production with six chips delivering 3.6 exaflops and 10x lower token costs for H2 2026 cloud rollouts.


Anthropic hired Clive Chan, OpenAI's second chip engineer, to build custom AI silicon and signal Nvidia independence ahead of its IPO.


Microsoft's Majorana 2 chip claims 1,000x more reliable qubits with a 20-second lifetime, targeting a commercial quantum computer by 2029.


Nvidia's FOX factory agent stack debuts at GTC Taipei, with Foxconn's MoMClaw slashing root-cause analysis 80% and machine failures 10%.


OpenAI's reasoning model disproved Erdos' 80-year geometry conjecture, proving AI can now solve original math that stumped humans for decades.


ChatGPT crossed 1 billion monthly users in May 2026, becoming the fastest app ever. But Claude is growing 10x faster, reshaping who wins the AI race.
Intel launched Xeon 6 Plus on its new 18A chip node with 288 cores, 2.5x performance gains, targeting agentic AI orchestration in data centers.
BYD confirmed humanoid robot development under codename Yao-Shun-Yu, deploying 20,000 units in its own factories in 2026 via Xi'an's 50,000-unit plant.
WeRide and Uber announced Spain's first commercial robotaxi pilot in Madrid, a 15-city European expansion with rides via the Uber app.
Agility Robotics' Digit became the first humanoid paid by the hour for real warehouse work, moving 100,000+ totes for GXO and Spanx.
Anthropic expands Project Glasswing to 150 organizations in 15+ countries, giving Claude Mythos vulnerability scanning to power, water, and health firms
Veeam's survey of 600 executives found 88% use AI agents but only 7% are ready, with 95% reporting unauthorized AI and just 25% offering alternatives.
Cambridge and DIOSynVax dosed 39 volunteers with the first AI-designed vaccine, covering the whole coronavirus family with one computed super-antigen.
Intel launched OpenVINO Physical AI, an open-source robotics runtime for vision-language-action models, to own edge robot deployment before Nvidia can.
Nvidia's State of AI 2026 survey of 3,200 firms finds 86% raising AI budgets and 88% reporting revenue gains, with finance and retail leading the returns.
MegaTrain trains 100B parameter LLMs on a single GPU at full precision for about $35K versus $200K, breaking the cluster barrier to frontier AI.
Google TurboQuant compresses the LLM KV cache 6x at 3 bits per value with no retraining, cutting AI inference memory and unlocking cheaper long context.
GitHub Copilot switched to token-based AI Credit billing on June 1, 2026, and heavy agentic users report bills jumping from $29 to $750 a month.
Anthropic reveals Claude now writes 80% of its production code, with task length doubling every four months. The recursive self-improvement loop has begun.


CrowdStrike beat Q1 estimates with $1.39 billion revenue and record cash flow, but in-line guidance and a 4-for-1 split sent the shares down 13%.


Broadcom AI chip revenue surged 143% to $10.8 billion in fiscal Q2 2026 and Q3 guidance points to $16 billion, yet the stock still fell on the news.


Marvell stock jumped 32.5% to a record high after Nvidia CEO Jensen Huang called it the next trillion dollar company at Computex 2026 in Taipei.


Walmart caps employee tokens on its Code Puppy AI agent as adoption outpaces returns, with 40% of firms reporting under 10% savings.


SoftBank will invest up to 75 billion euros to build 5 gigawatts of AI data centers in France by 2031, its largest European infrastructure bet ever.


Sysdig documented the first autonomous AI agent intrusion in the wild, an LLM that breached a Python notebook and stole an AWS database in under an hour.