Worldwide spending on artificial intelligence is expected to reach $2.67 trillion in 2026, a 49.5% increase over 2025, according to Gartner’s latest forecast. Infrastructure will account for more than half of that outlay, while growth will also extend to software, services, cybersecurity, and AI agent-based applications.
AI spending in 20 seconds
- Gartner puts worldwide AI spending at $2.67 trillion for 2026, up from $1.79 trillion in 2025.
- AI infrastructure will absorb $1.48 trillion, the largest slice of the market.
- AI software spending will reach $461,637 million and services will hit $576,481 million.
- AI agents and assistants will grow to $29,219 million in 2026.
- Gartner expects the global AI market to reach $3.64 trillion in 2027.
Gartner’s new estimate reflects how the expansion of artificial intelligence keeps shifting from models and early testing toward the infrastructure needed to run workloads at scale. Of the $2.67 trillion projected for 2026, $1.48 trillion will go to AI infrastructure.
According to Gartner, this category includes AI-optimized infrastructure as a service (IaaS), servers built for these workloads, AI-specific networking, semiconductors, and processing devices. The figure represents around 56% of the market’s total projected AI spending for the year.
Distinguished Analyst and Gartner Vice President John-David Lovelock believes demand for this infrastructure remains strong despite rising prices tied to certain components, especially memory. In his assessment, hyperscalers and service providers buying AI-optimized servers will continue to account for a significant share of spending.
The forecast puts AI infrastructure spending at $981,920 million in 2025. By 2027, Gartner expects it to climb to $1.98 trillion. The jump between those two years shows where much of the investment needed to expand data center processing capacity is concentrated.
Infrastructure dominates a market that keeps growing
Gartner’s table shows an AI market that keeps getting bigger. Total spending would go from $1.79 trillion in 2025 to $2.67 trillion in 2026 and $3.64 trillion in 2027 — a trajectory that echoes Gartner’s broader forecast that AI is driving overall global tech spending up to $6.37 trillion.
| Market | 2025 | 2026 | 2027 |
|---|---|---|---|
| AI services | $434,046M | $576,481M | $745,655M |
| AI cybersecurity | $25,920M | $51,347M | $85,997M |
| AI software | $288,168M | $461,637M | $656,353M |
| AI agents and assistants | $16,481M | $29,219M | $65,472M |
| Generative AI models | $13,021M | $28,266M | $51,620M |
| AI platforms for data science and machine learning | $19,405M | $26,445M | $35,552M |
| AI application development platforms | $6,885M | $9,541M | $12,478M |
| AI data | $826M | $3,126M | $6,480M |
| AI infrastructure | $981,920M | $1,484,397M | $1,977,685M |
| Total AI spending | $1,786,671M | $2,670,460M | $3,637,292M |
AI services will be the second-largest category in 2026, at $576,481 million, while software will reach $461,637 million. Gartner also expects a sharp rise in AI-driven cybersecurity spending, from $25,920 million in 2025 to $51,347 million in 2026.
Agents and assistants will also grow, though from a smaller base. Gartner estimates spending of $29,219 million in 2026 and $65,472 million in 2027.
The forecast reflects a shift in how companies are bringing AI into their systems. Gartner notes that software vendors are embedding agentic AI capabilities into existing products to meet demand for automating tasks and workflows.
The term agentic AI refers to systems capable of carrying out sequences of actions to complete specific goals, rather than simply responding to a query. In the scenario Gartner describes, these capabilities are being progressively built into the enterprise applications organizations already use.
Companies are leaning more on their software vendors
Gartner also observes a shift in the type of services companies are demanding. According to Lovelock, companies are turning less often to outside providers for large-scale business transformation projects and more often for narrower projects aimed at getting value out of the AI features already built into their software.
The firm estimates that the combination of transformation projects and smaller, more targeted projects could create a $1.2 trillion opportunity in AI services by 2030.
The forecast has also shifted in some segments compared with the previous quarter. Gartner raised its 2026 growth estimate for AI application development platforms from 28% to 39%. The firm ties this revision to growing interest among companies, software makers, and services firms in building AI applications tailored to specific needs.
The forecast for generative AI models has also increased. Gartner now expects 117% growth in 2026, up from its earlier estimate of 110%.
The firm also points to mounting pressure to cut the cost of running models and better tailor them to enterprise use cases. This is where domain-specific language models, or DSLMs, come in — a still-small but growing opportunity, according to Gartner.
The evolution of spending, then, isn’t limited to buying more compute capacity. It also spans application development, the services needed to integrate them, system protection, and the models used for specific tasks — a pattern also visible in Gartner’s outlook for the broader semiconductor market, which it expects to hit $1.555 trillion in 2026 on the back of AI demand.
Gartner also notes a change in its own forecasting methodology. In the current estimate, it has separated multifunction agents and assistants from AI software and added consumer-facing agents and assistants as their own category to reflect this emerging segment separately.
The $2.67 trillion figure should be understood as a spending forecast, not a result already on the books. The expected growth in the years ahead depends on enterprise demand, data center capacity, component prices, and the adoption of new AI applications.
The clearest takeaway is the sheer scale of infrastructure within the market. Gartner expects servers, networking, semiconductors, infrastructure services, and other components tied to AI processing to absorb $1.48 trillion in 2026. At the same time, software and services are growing their share as companies weave AI into the applications and processes they already use.
Frequently Asked Questions
How much will be spent on artificial intelligence in 2026?
Gartner forecasts worldwide spending of $2.67 trillion in 2026, up from $1.79 trillion in 2025 — a year-over-year increase of 49.5%.
Which segment will account for the most AI spending?
AI infrastructure will be the largest category, with $1.48 trillion projected for 2026. It includes servers, networking, semiconductors, IaaS, and other systems built for AI workloads.
How much will spending on AI agents and assistants grow?
Gartner expects spending on AI agents and assistants to reach $29,219 million in 2026 and $65,472 million in 2027.
What total AI spending does Gartner forecast for 2027?
The firm estimates worldwide spending of $3.64 trillion in 2027, up from the $2.67 trillion projected for 2026.
