Bill Ackman says this is a good environment to deploy capital. We’re still in the earliest innings of the AI economy and what may become the largest industrial buildout in human history. https://t.co/lC1du64C2d
)](https://ourworldindata.org/grapher/exponential-growth-of-computation-in-the-training-of-notable-ai-systems.png)
This infographic plots total training computation (petaFLOPs) for notable AI systems over time, showing the exponential jump in compute used by modern models — visual evidence of rapidly growing demand for GPUs, data centers, and power that supports the claim of an early-stage, large industrial buildout in the AI economy. ([ourworldindata.org](https://ourworldindata.org/grapher/exponential-growth-of-computation-in-the-training-of-notable-ai-systems))
Source: Our World in Data (Global Change Data Lab)
Research Brief
What our analysis found
Billionaire investor Bill Ackman is making a massive bet on the AI economy, with his $15.5 billion Pershing Square hedge fund allocating over half its portfolio to AI-linked stocks including Uber Technologies, Amazon, Alphabet, and Meta Platforms. His conviction that we are in the "earliest innings" of what could become "the largest industrial buildout in human history" is backed by staggering capital expenditure plans from tech giants: Alphabet is planning $175–$185 billion in 2026 capex, Amazon has outlined $200 billion, and Meta plans to spend $115–$135 billion, overwhelmingly directed toward AI infrastructure.
The numbers underpinning the broader AI market lend weight to Ackman's thesis. The global AI market was valued at an estimated $390.91 billion in 2025 and is projected to balloon to $3.5 trillion by 2033, growing at a compound annual growth rate of 30.6%. Private investment in generative AI alone surged to $33.9 billion in 2024, an 8.5x increase over 2022 levels. The United States dominates with $109.1 billion in private AI investment in 2024, dwarfing China's $9.3 billion and the U.K.'s $4.5 billion.
However, not everyone shares Ackman's optimism. Critics point to growing parallels with the dot-com era, noting that the S&P 500's cyclically adjusted price-to-earnings ratio has reached levels reminiscent of that bubble's peak. An MIT Media Lab report from August 2025 found that 95% of organizations are getting zero return from generative AI, raising serious questions about whether the enormous capital being deployed will translate into commensurate profits anytime soon.
Fact Check
Evidence from both sides
Supporting Evidence
Explosive AI market growth projections
The global AI market is estimated at $390.91 billion in 2025 and projected to reach $3.5 trillion by 2033, reflecting a CAGR of 30.6%, supporting Ackman's view that we are in the early stages of a massive economic shift.
Unprecedented corporate capital expenditure
Alphabet, Amazon, and Meta collectively plan to spend over $490 billion in 2026 capex alone, primarily on AI infrastructure, substantiating the claim of a historic industrial buildout.
Surging enterprise AI adoption
Organizational AI adoption jumped from 55% in 2023 to 78% in 2024, and generative AI usage in at least one business function more than doubled from 33% to 71% in the same period, indicating broad-based demand.
Massive private capital flows into AI
U.S. private AI investment reached $109.1 billion in 2024, and generative AI private investment hit $33.9 billion, an 18.7% year-over-year increase and 8.5 times the 2022 level.
PwC's macroeconomic estimate
PwC analysts project AI could generate $15.7 trillion in global economic value by 2030, reinforcing the scale of opportunity Ackman describes.
Strategic mega-deals continue
Google invested $40 billion in AI startup Anthropic in April 2026, OpenAI raised $122 billion reaching an $852 billion valuation in March 2026, and Microsoft committed $17.5 billion to AI infrastructure in India, all signaling sustained conviction among industry leaders.
Contradicting Evidence
AI bubble comparisons gaining traction
The S&P 500's cyclically adjusted price-to-earnings (CAPE) ratio has reached levels similar to the peak of the dot-com bubble, and analysts have raised concerns that a circular flow of investments among leading AI firms may be artificially inflating valuations.
Bank of England overvaluation warning
The Bank of England has flagged risks of a global market correction due to possible overvaluation of leading AI technology firms, suggesting that current prices may not be sustainable.
Negligible enterprise returns so far
An MIT Media Lab report from August 2025 found that 95% of organizations are getting zero return from generative AI, and many AI initiatives struggle to move beyond pilot phases or deliver measurable improvements at scale.
Exponentially rising training costs
The cost of training a single frontier AI model has surged from roughly $1,000 in 2017 to nearly $200 million in 2024, with projections suggesting costs could reach billions by 2030, threatening the economics of continued buildout.
Revenue justification gap
Critics argue that despite enormous capital inflows, very few AI companies are generating significant profits, raising the concern that the investment thesis rests more on sentiment and future expectations than on current fundamentals.
Report an Issue
Found something wrong with this article? Let us know and we'll look into it.