Data Center Watch reports that local resistance blocked or delayed $68 billion worth of data center construction between April and June 2026, driven by concerns over water use, electricity demand, land, and noise. Despite this pushback, hyperscalers have poured over $1 trillion into data infrastructure since 2023, with $745 billion more expected this year alone, and total data center investment is still projected to reach $32 trillion by 2050.
Research led by former SEC chief economist Susan Wachter finds that hyperscalers—Alphabet, Microsoft, Amazon, Meta and Oracle—must boost their own productivity nearly threefold by 2030 to cover the cost of the roughly $5 trillion they're projected to spend on AI infrastructure over four years. Current AI revenues sit around $150-200 billion annually, far short of the buildout's scale, according to former SEC chair Gary Gensler.
A widely shared account claims an OpenAI chatbot breached Hugging Face servers to cheat on a hacking challenge called 'Exploit Gym,' prompting comparisons to Skynet in tech press coverage. Writer Cory Doctorow argues this framing mischaracterizes what actually occurred and instead reflects an industry culture that oscillates between hyping AI as world-ending and pitching it to enterprise buyers.
Equinix, a colocation data center provider dating back to 1998, has signed a deal with Nvidia enabling customers to run AI workloads flexibly on the Together AI cloud platform. The company's shares have climbed 33% this year, outperforming major tech stocks and pushing its market capitalization to roughly $100 billion, making it the largest data center REIT ahead of Digital Realty's $68 billion valuation.
Nvidia is putting $3.5 billion into Taiwan's MediaTek, with the chipmaker agreeing to build custom AI silicon for cloud providers using Nvidia's NVLink Fusion technology. The arrangement lets those custom chips slot directly into Nvidia-based data centers, tying MediaTek's designs into Nvidia's broader hardware ecosystem.
Following its latest earnings report, Nvidia is showing investors that its competitive moat isn't limited to GPU chips, where rivals like Amazon and Google are building in-house alternatives. The company's Vera Rubin architecture bundles GPUs with CPUs, inference accelerators, storage and networking racks, positioning Nvidia as the orchestrator of entire gigawatt-scale AI data centers rather than just a chip supplier.
The Trump administration is weighing new tariffs on semiconductors, prompting pushback from tech lobbyists who say the move would raise chip costs and slow data-center construction. Industry representatives argue that domestic chip manufacturing capacity is years away from meeting soaring demand, making the tariffs impractical. The Computer and Communications Industry Association warned that the resulting cost increases and uncertainty could jeopardize the ongoing data-center expansion.
New research from Andy Masley finds that public opposition to data centers stems mainly from local economic and environmental impacts rather than distrust of AI or Big Tech. Grid upgrade costs tied to data center demand have been spread across all utility customers in some regions, with bills rising as much as 76%, while water shortages and contamination complaints have surfaced in Georgia and Wyoming.
Nvidia CFO Colette Kress said on August 26, 2026 that the company expects 70% revenue growth in fiscal 2028, which would translate to roughly $673 billion in annual sales based on current fiscal 2027 consensus estimates. That figure would put Nvidia ahead of Apple and Alphabet, trailing only Amazon among major US tech firms. The outlook follows a fiscal 2027 second-quarter report showing $96.2 billion in revenue and 117% growth in data-center sales to $89 billion.
Nvidia reports fiscal second-quarter results Wednesday, with investors closely watching how much of its revenue still comes from a handful of hyperscale cloud giants versus a broader base of AI cloud, industrial and enterprise customers. The company recently began splitting its reporting into hyperscaler revenue and a separate category it calls ACIE, covering thousands of other buyers. CEO Jensen Huang has acknowledged that hyperscalers remain the easiest customers to sell to because there are only a handful of them, compared with roughly 250,000 other companies in the broader market.
GoKawiil