Money is moving into artificial intelligence (AI) faster than into almost any technology before it. Gartner expects global AI spending to reach $2.6 trillion in 2026, up 47% from the year before.
The total depends on what is counted. Some estimates cover only chips and data centers. Others add software, services, venture capital and company buyouts.
This article uses data from Reuters, Goldman Sachs, Stanford University, the International Monetary Fund (IMF) and a 2026 funding review. It explains each measure, compares it with past tech booms and looks at the risks.
Key AI Investment Numbers for 2026
There is no single agreed figure for AI spending. Each group measures something different, so the numbers should not be added together.
Gartner puts worldwide AI spending at $2.6 trillion in 2026. That covers hardware, software, services and business purchases. The forecast was raised from $2.52 trillion in January.
Goldman Sachs Research counts only investment, and it estimates $1 trillion worldwide in 2026. About $581 billion of that is in the United States. Goldman says the common Big Tech spending forecast of $794 billion understates global AI investment by about $200 billion.
Goldman also estimates that total AI investment since 2022 will reach $1.8 trillion by the end of 2026.
Inside the Gartner total, AI-optimized servers are growing 49% this year and make up about 17% of spending. Agentic AI software, which handles tasks on its own, is on pace to grow 141% to nearly $202 billion.
Gartner says total AI spending could reach $3.5 trillion by 2027. All technology spending is projected at $5.6 trillion in 2026, up 7.8%.
| Measure | Amount | What it covers |
|---|---|---|
| Total AI spending, 2026 (Gartner) | $2.6 trillion | Hardware, software, services |
| Global AI investment, 2026 (Goldman Sachs) | $1 trillion | Capital spending tied to AI |
| Big Tech capex, 2026 | $725 billion | Amazon, Microsoft, Alphabet, Meta |
| Corporate AI investment, 2025 | $581.7 billion | Venture capital, private equity, M&A |
How AI Spending Compares With Past Big Projects
Reuters compared AI with several famous projects, using today’s dollars. The Manhattan Project cost about $30 billion. The Apollo program cost about $298 billion.
Investors put nearly $1.6 trillion into AI between 2013 and 2024, according to Stanford data used by Reuters. That is more than five times the cost of Apollo.
Past technology booms were larger in total. Companies spent about $910 billion on telecom networks from 1996 to 2000. Railroad booms in the 1840s and 1860s drew about $3 trillion, based on research by Andrew Odlyzko of the University of Minnesota.
Goldman Sachs says AI spending as a share of the economy is still within the range of earlier technology cycles. Those cycles peaked at 2% to 5% of gross domestic product (GDP).
The speed is what stands out. Reuters said AI has pulled in money at a faster rate than past booms, and in a shorter time.
The 2026 funding review adds a venture capital comparison. The peak quarter of the dot-com boom drew about $28 billion in venture funding in the second quarter of 2000. The first quarter of 2026 drew about $300 billion, roughly ten times as much.
Reuters also noted that forecasts for 2025 added another $375 billion, so a single year of AI investment would top the full cost of the Apollo program.
The Bank for International Settlements warned in its 2026 annual report that AI has echoes of canal, railway and dot-com booms. It said new technology can change the economy while still leaving investors exposed if prices run ahead of results.
Corporate and Private AI Investment
Stanford’s 2025 AI Index reported that corporate AI investment reached $252.3 billion in 2024. Private investment rose 44.5% that year, and mergers and acquisitions rose 12.1%. Total investment has grown more than thirteen times since 2014.
The 2025 numbers were much higher. Global corporate AI investment hit $581.7 billion, more than double the 2024 total, based on the 2026 funding review. It passed the earlier record of $360 billion from 2021.
The 2021 peak was driven by company buyouts. The 2025 peak was driven by private investment, which grew 127.5% to $344.7 billion.
Generative AI took nearly half of all private AI funding in 2025, growing more than 200% from a year earlier. In 2024 it stood at $33.9 billion, up 18.7% from 2023.
Billion-dollar funding events rose from 15 to 28 in 2025. The number of newly funded AI companies grew 71%.
Reuters reported that AI startups raised over $70 billion worldwide in the first quarter of 2025. That was nearly 60% of all venture capital financing in the period.
Use of the technology is growing as fast as the funding. Stanford says generative AI reached 53% of the population within three years, faster than the personal computer or the internet.
Stanford estimates that generative AI tools were worth $172 billion a year to U.S. consumers by early 2026. Many of those tools are free or low cost.
Much of the money goes to physical assets. Stanford data shows companies spent $37 billion on AI infrastructure in 2024. More than 500 large data centers were built between 2021 and 2024, and McKinsey estimates the world will need $5.2 trillion of data center investment by 2030.
AI Venture Capital Statistics
Venture capital set new records in 2026. About $300 billion went into roughly 6,000 startups in the first quarter. A second quarter added $210 billion, bringing the first half to $510 billion.
That is more than the $440 billion invested in all of 2025. The old half-year record was $375 billion in the second half of 2021.
AI took about 80% of all venture funding in the first quarter of 2026, up from 55% a year earlier. Roughly $242 billion went to AI companies.
The money is also bunched in one place. U.S. companies raised $250 billion in the first quarter, or 83% of the global total. The San Francisco Bay Area alone captured $122 billion, about 76% of U.S. AI funding.
A few very large rounds drove the total. OpenAI and Anthropic together took $217 billion in the first half, or 43% of all venture dollars worldwide. Without the mega-rounds, activity looked close to 2024 and 2025 levels.
| Company | Q1 2026 round |
|---|---|
| OpenAI | $122 billion |
| Anthropic | $30 billion |
| xAI | $20 billion |
| Waymo | $16 billion |
| Total | $188 billion (about 63% of Q1 venture funding) |
The exit market also opened up. Thirty-two companies went public at valuations above $1 billion in the second quarter. Mergers and acquisitions hit a quarterly record, with 24 deals worth $113 billion.
Another tracker that counts sovereign wealth funds put the first quarter at $330.9 billion across 8,464 deals. The gap comes from method, not from a conflict in the data.
Big Tech Capital Spending
Amazon, Microsoft, Alphabet and Meta raised their 2026 capital spending plans to about $725 billion. That is up 77% from about $410 billion in 2025 and roughly triple the $226 billion spent in 2024.
Forecasts say the group could pass $1 trillion in 2027. Cumulative spending from fiscal 2025 to 2030 is estimated at $5.3 trillion. Most of it goes to AI computing, custom chips, data centers and power.
| Company | 2026 capex plan | Detail |
|---|---|---|
| Amazon | About $200 billion | $44.2 billion spent in Q1; AWS revenue up 28% |
| Microsoft | About $190 billion | About $25 billion tied to higher memory and parts costs |
| Alphabet | $180 to $190 billion | Google Cloud revenue up 63% to $20 billion in Q1 |
| Meta | $125 to $145 billion | Raised from $115 to $135 billion on memory chip prices |
Each company said demand is higher than the supply it can build. Microsoft expects to stay short of capacity through at least the end of 2026. Google Cloud’s backlog reached $462 billion.
Wall Street has noticed. Reuters reported in December 2025 that the “Magnificent Seven” stocks had beaten the rest of the S&P 500 by nearly 15 percentage points since January 2025.
AI Company Valuations
Private AI firms now carry some of the highest values in the world. The 2026 funding review lists Anthropic at $965 billion after a $65 billion round on May 28, 2026. OpenAI is listed at $852 billion after its $122 billion raise.
Anthropic’s run-rate revenue crossed $47 billion in May, up from $10 billion at the start of 2025, according to the review. It also reported that Anthropic filed confidentially for a U.S. stock listing on June 1, targeting October.
Other firms are growing quickly from a smaller base. Databricks was valued at $188 billion in July, up 40% in five months, with annual revenue of $5.4 billion. SpaceX agreed to buy Cursor for $60 billion in stock, and Cursor’s annual revenue went from $100 million in January 2025 to $2 billion by February 2026.
China’s DeepSeek raised about $7.4 billion in June at a value above $52 billion. It was its first outside funding round.
Chip makers have gained too. Reuters reported that Nvidia’s value rose more than ten times after ChatGPT launched, to about $4.5 trillion in late 2025.
Some early-stage firms have high values with little or no revenue. Thinking Machines Lab reached $50 billion after a $5 billion round in March 2026. Safe Superintelligence is valued at $32 billion with no product on the market.
Figure AI, a humanoid robot maker, reached $39 billion. Perplexity is valued at about $21 billion with annual revenue near $450 million, and Mistral AI of France is valued at about $15 billion.
AI Investment by Country
The United States leads by a wide margin. Stanford’s 2026 AI Index says U.S. private AI investment reached $285.9 billion in 2025, more than 23 times China’s $12.4 billion. The U.S. also had 1,953 newly funded AI companies, more than ten times the next country.
Stanford counts 5,427 data centers in the U.S., more than ten times any other country. The 2026 funding review gives a lower count of 4,011 as of March 2026, so the totals depend on the source.
China’s private figure leaves out government money. The funding review says about $184 billion in government guidance funds has gone into AI firms since 2000.
| Measure | United States | China |
|---|---|---|
| Private AI investment, 2025 | $285.9 billion | $12.4 billion |
| Private AI investment, 2024 | $109.1 billion | $9.3 billion |
| Big tech capex, 2025 | More than $400 billion | About $63 billion |
| Model performance gap, March 2026 | 2.7% in favor of the top U.S. model | |
Spending is not limited to those two. France committed 109 billion euros, about $120 billion, to a national AI plan. Saudi Arabia launched a $100 billion program called Project Transcendence.
Use of AI tools varies by country. Stanford reports adoption of 64% in the United Arab Emirates and 61% in Singapore. The U.S. ranks 24th at 28.3%.
Stanford also found that the number of AI researchers and developers moving to the U.S. has dropped 89% since 2017. Among surveyed countries, the U.S. reported the lowest trust in its own government to regulate AI, at 31%.
The U.S. is on track to spend about 2% of GDP on AI and data centers in 2026. Norway and Saudi Arabia follow at about 0.7%, China at 0.4%, the euro zone at 0.2% and Canada at 0.15%.
Returns, Risks and Bubble Concerns
Spending is running ahead of proven returns. In the funding review, 56% of CEOs said AI had produced neither more revenue nor lower costs in the past year. Only 12% reported both.
About 88% of organizations use AI in at least one function, but only 39% report a profit impact at the company level. That is a 49-point gap.
Stanford’s 2025 report found that most firms with gains see small ones. Cost savings were usually under 10% and revenue gains usually under 5%.
Other studies point the same way. An estimated 95% of generative AI pilots at large companies failed to deliver meaningful returns, according to the funding review. It describes the cause as a gap between what tools can do and how ready the organization is.
The review also notes that U.S. AI revenue is about $12 billion a year, compared with more than $500 billion in infrastructure spending.
| Business function | Share reporting cost savings | Share reporting revenue gains |
|---|---|---|
| Service operations | 49% | 57% |
| Supply chain management | 43% | 63% |
| Software engineering | 41% | Not reported |
| Marketing and sales | Not reported | 71% |
Reuters also raised concerns about circular deals. Nvidia pledged up to $100 billion to OpenAI, and much of it is expected to return as chip purchases. OpenAI signed a $300 billion computing contract with Oracle over about five years.
Oracle was reported to be taking on another $38 billion in debt for AI, on top of more than $100 billion it already owed. Andrew Odlyzko, who studies financial bubbles, called such deals a warning sign about real customer revenue.
There are counterpoints. The top tech firms have net margins above 25%, compared with 13% for the S&P 500 average. Much of the spending is paid for by profitable companies rather than by debt.
Carnegie Mellon finance professor Bryan Routledge told Reuters that stock prices look forward, so they do not need positive cash flow today to reflect value.
AI Investment Outlook and Economic Impact
Goldman Sachs expects AI investment to keep rising as a share of the economy. It projects U.S. AI capex at 1.8% of GDP in 2026, 2.5% in 2027 and 2.8% in 2028. Global investment would reach 1.4% of world GDP by 2028.
Goldman said all of its leading indicators sit near the top of their range since 2022. It reads that as a strong near-term outlook for spending.
The IMF said on October 13, 2026 that AI investment is helping lift world growth. It forecasts global growth of 3.1% in 2026 and 3.4% in 2027, even with oil above $100 a barrel from the war in the Middle East.
The gains are uneven. Just 32 countries now account for 95% of global growth, down from 54 in April. Countries that supply AI hardware, such as South Korea, Malaysia, Mexico and Thailand, saw upgrades.
The IMF expects U.S. growth of 2.3% in both 2026 and 2027, and said the AI boost is offset by the large share of AI equipment that is imported. Euro zone growth was raised to 1.1% for 2026, and Britain’s to 1.3%, both partly because of AI investment.
China is forecast at 4.5% and India at 7.0%. The IMF projects global consumer price inflation of 4.7% in 2026 and said the AI investment surge keeps upward pressure on prices.
| IMF scenario | Result |
|---|---|
| Baseline forecast | Global growth of 3.1% in 2026 and 3.4% in 2027 |
| Faster AI adoption and investment | Global GDP up a further 0.6% by 2029 |
| AI productivity gains fail to appear | U.S. GDP down 0.8% directly; global output down 1.1% and U.S. output down 1.8% by 2028 |
The IMF listed a reassessment of AI profits as one of its major downside risks. A sharp drop in investment would lower asset prices and tighten financial conditions.
Enterprises face a test in the second half of 2026. Forecasters in the funding review expect companies to ask for measurable returns, and they estimate that 40% of AI projects started in 2024 and 2025 could lose funding.
What the AI Investment Data Shows
The numbers point in one direction. Global AI spending is at a level that matches or passes earlier technology booms, and the largest firms plan to spend more in 2027.
The money is also concentrated. A few companies, mainly in the United States, take most of the venture funding and capital spending.
The open question is return. Adoption is high, but most businesses report small gains so far. The next year of results will show whether the spending pays off.
Sources
- Investment in AI is exploding (Reuters Graphics, Dec. 5, 2025)
- Global AI Investment Is Forecast to Exceed $1 Trillion in 2026 (Goldman Sachs, Aug. 7, 2026)
- The 2026 AI Index Report: Top Takeaways (Stanford HAI)
- AI investment props up global growth, but war and debt threaten outlook, IMF says (Reuters, Oct. 13, 2026)
- The 2025 AI Index Report: Top Takeaways (Stanford HAI)
- AI Global Funding Statistics 2026: The Complete Data Breakdown (Vincee Cole, Aug. 4, 2026)
