Anthropic's Claude model has reportedly secured 14.8 gigawatts of compute, involving $517 billion in investment. This figure highlights the massive capital commitments in AI, sparking debates about an investment bubble.
In Q2 2026, Chinese tech giants sharply increased AI capital expenditure, driven by inference demand and chip availability. However, Jefferies sees greater sustainability risks in US cloud spending, highlighting rising capex relative to revenue and the challenge of monetising AI infrastructure investments.
PwC projects $31.6 trillion in AI data center spending through 2050, but power availability decides where the capital lands. The US captures 48%.
Goldman Sachs traders' report focuses on Europe's yield test and the spillover effects of AI capex, suggesting potential downsides to the AI boom. It implies massive AI investment could cause market volatility but provides no specific figures.
A new report by Moody's Ratings suggests that the massive gap in AI spending between US and Chinese tech giants may not buy the advantage expected for American giants, as lower domestic costs and heavy state support allow Chinese firms to secure far more computing power per dollar. While US hyperscalers outspent their Chinese counterparts by a staggering margin, the physical gap in computing capacity was nowhere near as wide as those mega-budgets suggested.
According to SEMI's report, global semiconductor equipment shipments reached $40.53 billion in Q2 2026, up 23% year-over-year and 11% quarter-over-quarter, marking a second consecutive record high. The SEMI president attributed the growth to AI infrastructure demand, with increases across Asia, North America, and Europe, highlighting the global nature of the demand.
At the G20, Nvidia CEO Jensen Huang elaborated on the massive scale of AI infrastructure investment, stating that a 1GW computing center could cost $60 billion. He emphasized the economic logic behind such investment, attempting to justify the high AI capital expenditure.
HyperVault, a subsidiary of Tata Consultancy Services (TCS), announced it has acquired 264 acres of land in Hyderabad, India, to build a large AI data center campus with a capacity of up to 1GW. The company and its partners expect to invest up to 700 billion rupees (about $7.4 billion).
Tata Consultancy Services announced that its subsidiary HyperVault will build a 1-gigawatt AI data center campus in Hyderabad, India, with an estimated investment of up to 700 billion Indian rupees (approximately 49.787 billion yuan). The campus will support high-density GPU deployment for AI model training and inference, and will be built in phases to match customer demand. The project is expected to create thousands of jobs and become one of India's largest AI infrastructures.
A report notes that India's cyclical recovery faces margin risks from an oil price shock and slowing AI capex. Rising oil prices could increase costs, while a slowdown in global AI investment may affect demand for India's tech-related sectors.
According to projections, cumulative global AI capital expenditure will reach $31.6 trillion by 2050, surpassing the cumulative use of traditional building materials like concrete. This forecast highlights the massive scale of AI infrastructure investment, potentially raising concerns about a bubble.
According to ET Datacenters, global AI capital expenditure is projected to reach $31.6 trillion by 2050, highlighting the massive scale of AI infrastructure investment. The report also notes that chips will overtake concrete in usage, symbolizing the dominance of AI hardware in construction and infrastructure.
Enterprise AI spending appears to be accelerating at a substantial pace. Gartner forecasts worldwide AI spending will reach approximately $2.59 trillion in 2026, a 47% increase from the previous year. Much of that investment is expected to flow toward infrastructure capable of supporting increasingly sophisticated models and AI-enabled workflows.
According to capacityglobal.com, global AI capex is projected to reach $31.6 trillion by 2050, with chip investment surpassing traditional infrastructure like concrete. This forecast highlights the long-term growth trend in AI infrastructure investment.
The article discusses the AI capex boom in 2026 and its impact on prediction markets, but does not provide specific figures.
This article breaks down the cost structure of AI infrastructure investment, tracing where each $100 of AI capex goes, covering chips, servers, data centers, and more. It aims to help investors understand the multiplier effects and potential risks of AI investment.
ByteDance reportedly secured a $29.6 billion loan for general corporate purposes and plans to raise its 2026 capital expenditure to as much as $70 billion, mainly for AI infrastructure. OpenAI released GPT-6 Astra, positioned as an AI agent. Nvidia agreed to acquire Hugging Face for $12.93 billion, pledging to keep it an open platform.
While Nvidia is synonymous with the AI boom, a lesser-known group of power and cooling equipment suppliers is cashing in on a global data centre construction spree as developers race to avoid infrastructure bottlenecks. Energy-hungry data centres have triggered a surge in demand for equipment ranging from transformers to advanced cooling systems, creating winners across Asia's supply chain, though earlier stock-price gains have moderated. McKinsey forecasts nearly $7 trillion in data-centre investment globally by 2030; Nvidia last week said it expects AI spending to remain robust for years.
The article reports that global AI infrastructure investment is expected to reach $31 trillion, with data center capital expenditure continuing to exceed market expectations. This figure highlights the massive spending by tech giants in AI, fueling debates about an AI bubble.
The article analyzes the competition between two PC giants in AI computing infrastructure, as they bet on different technological routes to capture market growth. This reflects the continued heating up of AI hardware investment, but no specific financial figures are provided.