AI Spending 2026: Gartner Forecasts $2.7 Trillion — What It Means for Your Business

Every business leader is being asked the same question this year: how much should we be spending on AI, and on what? The latest answer from the market is striking. On 16 September 2026, Gartner raised its forecast for AI spending in 2026 to $2.7 trillion worldwide, a 49.5% increase on 2025. Software is growing even faster than the headline, and spending on AI agents and assistants is climbing steepest of all.

Numbers that large can feel irrelevant to a startup, an SME or a mid-sized enterprise. They are not. They tell you where vendors are putting their products, where prices and talent will move, and which kinds of AI investment the market believes will pay back. In this analysis we break down Gartner's AI spending 2026 forecast, explain what is driving each segment, and turn it into practical guidance for planning your own AI budget, so you invest in results, not hype.

AI Spending 2026 at a Glance

Quick answer: Gartner forecasts worldwide AI spending of $2.7 trillion in 2026, up 49.5% year over year, and expects a further 36.2% rise in 2027. More than half of the money goes to AI infrastructure, but AI software (up 60.2%) and AI agents and assistants (up 77.3%) are the fastest-growing areas that businesses actually buy and use.

• Total: about $2.7 trillion in 2026, up 49.5%, according to Gartner's September 2026 forecast.
• Infrastructure (AI servers, chips, data centres) remains the largest area at roughly $1.48 trillion.
• AI software is forecast to grow 60.2% to about $462 billion as vendors embed AI in existing products.
• AI agents and assistants spending is forecast to rise 77.3%, the fastest of any segment.
• AI security spending is expected to almost double.
• Generative AI is in the "Trough of Disillusionment", so companies are favouring simpler, embedded AI features with clear operational value.
• Custom AI applications are a growing driver: Gartner raised its 2026 growth outlook for AI application development platforms from 28% to 39%.

How Much Will the World Spend on AI in 2026?

Gartner expects the world to spend about $2.7 trillion on AI in 2026, 49.5% more than in 2025. This is an upgrade on its earlier forecasts this year: in January Gartner put 2026 AI spending at around $2.5 trillion, and in May it forecast 47% growth. Each revision has moved upward as infrastructure orders, software adoption and enterprise projects have outpaced expectations.

The momentum is not expected to stop in 2026. In comments to CIO.com, Gartner Distinguished VP Analyst John-David Lovelock said AI spending should grow a further 36.2% in 2027. He also noted that the AI boom has not simply drained existing IT budgets: "there wasn't a diversion," he said, describing organisations receiving net new funding for AI, alongside a growing tendency to rebrand existing investments as AI-related.

To put the AI number in context, Gartner's July 2026 forecast put total worldwide IT spending at $6.37 trillion for 2026, up 14.2%. AI is not a side project inside IT budgets any more; it is one of the main forces shaping them. Lovelock's longer-term view is blunt: by 2030, "every dollar is going to be an AI dollar in one way or another," as AI becomes embedded in devices, enterprise software and business projects.

Where Is the AI Money Going? Gartner's Forecast by Segment

The headline figure blends very different kinds of spending, from hyperscale data centres to the AI features inside your accounting software. Gartner's segment breakdown shows where the growth actually is.

Source: Gartner forecast of worldwide AI spending, September 2026, US dollars (figures as published, rounded).

Three patterns stand out:

1. Infrastructure dominates the total. Gartner says capacity growth from hyperscalers buying AI-optimised servers will remain the largest spending area, and describes demand for AI infrastructure as strong and "inelastic" even as memory prices rise. Gartner also now expects semiconductor spending to reach about $2 trillion by 2030, driven largely by memory and storage for AI servers.
2. Software and services are where businesses spend. AI services (consulting, implementation and managed services) and AI software together are forecast at over $1 trillion in 2026. This is the part of the market most companies touch directly.
3. The smallest segments are growing fastest. AI agents, generative AI models, AI data and AI security are all forecast to roughly double or more between 2025 and 2026. They are small today, but they show where the next wave of products and budgets is heading.

For most organisations, the infrastructure numbers matter indirectly: they shape the price and availability of cloud AI services you rent. The software, services, agents and security lines are the ones that map directly onto your own roadmap.

Why Is AI Software Growing Faster Than the Market?

AI software is forecast to grow 60.2% in 2026 because almost every software vendor is now building AI into the products businesses already use. Gartner notes that vendors across software categories are "rapidly embedding agentic AI within their existing products to maintain relevance in the market and defend against new cross-functional agents."

In practice, that means your CRM, helpdesk, accounting, HR, ecommerce and productivity tools are shipping AI features, often as paid add-ons or premium tiers. Gartner observes that enterprises are using these simpler, embedded AI features from incumbent providers to improve operational efficiency, automate workflows, improve customer engagement and support better decisions.

For business owners, this has two consequences:

• Your software bills will change. Expect more AI add-ons, usage-based AI pricing and bundles. Budget for them deliberately rather than letting them creep in licence by licence.
• You may already own useful AI. Before commissioning anything new, audit the AI features already included in the tools you pay for. Some quick wins, such as summarising tickets, drafting emails or classifying documents, may be one configuration away.

Embedded AI has limits, though. It usually works inside one product. The biggest efficiency gains tend to come from AI that works across systems, such as reading an order from WhatsApp, checking stock in your ERP and updating your CRM. That requires integration, which is where custom work comes in.

AI Agents and Assistants: The Fastest-Growing Segment

Spending on AI agents and assistants is forecast to grow 77.3% in 2026, from about $16.5 billion to $29.2 billion, and to more than double again to about $65.5 billion in 2027. Gartner has also started tracking cross-functional and consumer agents as their own forecast category, a sign of how quickly the market is forming.

An AI agent is software that can take a goal, plan steps and use tools, such as your databases, APIs or email, to complete tasks with limited human input. An AI assistant is typically narrower: it answers questions or drafts content inside a workflow while a person stays in control.

Useful, realistic agent and assistant use cases for businesses in 2026 include:

• Customer support assistants that answer routine questions on your website or WhatsApp and hand complex cases to staff with full context.
• Sales and CRM assistants that qualify leads, log calls, update records and draft follow-ups.
• Back-office agents that extract data from invoices, purchase orders or forms and enter it into your ERP or accounting system for approval.
• Internal knowledge assistants that answer employee questions from your policies, manuals and past tickets.
• Operations agents that monitor orders, stock or service tickets and trigger follow-ups when something is late.

The growth numbers do not mean every agent project succeeds. Agents need clean data, well-defined permissions, reliable integrations and human approval for anything consequential. We cover adoption and ROI in more detail in our guide to enterprise AI agents in 2026, and costs in our AI chatbot development cost guide.

What Does "GenAI in the Trough of Disillusionment" Mean for Businesses?

Gartner places generative AI "firmly in the Trough of Disillusionment" in 2026. On Gartner's Hype Cycle, this is the stage after peak excitement, when early projects expose real-world limits and interest shifts from experiments to proven, practical value.

It does not mean spending is falling; the forecast clearly shows it is not. Gartner even raised its 2026 growth estimate for generative AI models from 110% to 117%. What it means is that expectations are becoming more realistic. Companies are asking harder questions about accuracy, cost, security and measurable return, and they are favouring focused use cases over broad "AI transformation" programmes.

For a business planning AI now, the trough is actually good news:

• Less pressure to chase hype. You can pick use cases on business merit rather than fear of missing out.
• Better-understood patterns. Retrieval-augmented generation, human-in-the-loop approvals and evaluation testing are now well-established ways to make generative AI reliable.
• More mature tools. Models, APIs and platforms have improved, and prices for many AI capabilities keep changing as competition increases.

The lesson is simple: treat generative AI like any other software investment. Define the problem, measure the baseline, pilot on real data, and scale only what proves itself.

Should You Build Custom AI, Buy a Tool or Use Embedded AI?

Most businesses need a mix of all three, and Gartner's forecast shows the market moving in each direction at once. Embedded AI is spreading through existing software, while Gartner raised its 2026 growth outlook for AI application development platforms from 28% to 39% because enterprises, software providers and services firms are building custom AI applications tailored to their needs.

A practical rule: use embedded AI for productivity, buy for commodity tasks, and build where AI touches your competitive advantage or connects several systems. Custom does not have to mean starting from scratch. Many of the most valuable projects combine a commercial model with your own data, business rules and integrations, delivered as a feature in your existing web or mobile app.

AI Security Spending Is Nearly Doubling — Why It Matters

Gartner forecasts AI cybersecurity spending to almost double in 2026, from about $25.9 billion to $51.3 billion. That reflects two realities: attackers are using AI, and every new AI system creates new risks that need protecting.

If you are adding AI to your business, build these safeguards in from the start:

• Data access controls: an AI assistant should only see the data the user asking it is allowed to see.
• Prompt injection defences: treat content from emails, web pages and documents as untrusted input to AI systems.
• Human approval for actions: payments, deletions, customer communications and record changes should be confirmed by a person, at least at first.
• Logging and monitoring: keep an audit trail of what the AI saw, suggested and did.
• Privacy and compliance checks: know where your data goes, which model providers process it, and whether that fits your obligations under laws such as GDPR or India's DPDP Act.

Security is not a reason to delay AI. It is a reason to implement it properly, with an experienced partner and a clear review process.

How Should Businesses Plan Their AI Budget for 2027?

The best AI budgets start from business problems, not technology. Lovelock's advice to CIOs is to expect "at least three major transitions" in the next two to three years and to make the "balance between risk and reward" the top priority. For a startup, SME or growing enterprise, that translates into a simple, staged plan:

1. Audit what you already pay for. List the AI features in your current software and switch on the useful ones before buying more.
2. Pick two or three high-value use cases. Look for repetitive, high-volume work with clear inputs and outputs, such as support queries, data entry, lead qualification or report preparation.
3. Measure the baseline. Record time spent, error rates, response times or conversion before you start, so you can prove the result.
4. Fix the data first. AI is only as good as the data and integrations behind it. Budget for cleaning, connecting and securing data; see our guide to AI-ready data.
5. Pilot in weeks, not months. Build a focused pilot on real data with a small group of users and human review.
6. Plan running costs. AI has ongoing costs: model usage, hosting, monitoring, updates and support. Include them in your business case.
7. Scale what works, stop what does not. Expand successful pilots to more teams or channels, and retire the ones that do not meet their targets.

Costs vary widely with scope, integrations and usage, so treat any single number with caution. A useful way to size an initial budget is to separate a fixed build cost for discovery, design, development and integration from a variable run cost for model usage, hosting and support, and to review both every quarter as usage grows. For automation opportunities that may not even need AI, see our business process automation guide for SMEs.

Where CodeBase Coders Fits: Turning AI Spending into Business Results

CodeBase Coders helps businesses turn ideas into scalable digital products and improve existing processes through software, automation, AI, integrations and modern web technologies. As AI budgets grow, our role is to make sure the money turns into working systems your team actually uses. We help with:

• AI development and integration: custom AI features, assistants and agents connected to your data and systems, through our AI solutions.
• AI chatbots and WhatsApp automation: support and sales assistants on your website and WhatsApp Business API, with handover to your team. Read our WhatsApp Business API guide.
• CRM, ERP and API integration: connecting AI to Salesforce, HubSpot, Zoho, ERPs and third-party APIs through our software integration services.
• Custom software, web and mobile apps: building AI into products your customers use, with our software development and mobile app development teams.
• Data readiness: pipelines, dashboards and data quality work through our data science and analytics services.
• Strategy and modernisation: AI roadmaps through technology consulting, and making older systems AI-ready with legacy application modernisation.
• Dedicated teams and support: dedicated development teams and ongoing maintenance so your AI keeps improving after launch.

Frequently Asked Questions

How much will the world spend on AI in 2026?

Gartner forecasts worldwide AI spending of about $2.7 trillion in 2026, up 49.5% from 2025, according to its September 2026 forecast. It expects spending to grow a further 36.2% in 2027.

Which area of AI spending is growing fastest in 2026?

AI agents and assistants are the fastest-growing segment in Gartner's forecast, with spending expected to rise 77.3% in 2026. AI software overall is forecast to grow 60.2%, and AI security spending is expected to almost double.

Why is AI infrastructure the largest part of AI spending?

Hyperscalers and cloud providers are buying AI-optimised servers, chips, networking and storage to run AI models, and Gartner describes this demand as strong and inelastic. Infrastructure accounts for roughly $1.48 trillion of the 2026 total, while businesses mainly access it indirectly through cloud AI services.

What does it mean that generative AI is in the Trough of Disillusionment?

It means expectations are resetting after the initial hype. Companies are focusing on practical, measurable uses of generative AI, often through AI features embedded in software they already use, rather than broad experimental programmes. Spending is still growing strongly.

Should small and medium businesses invest in AI in 2026?

Yes, if it is tied to a clear business problem. Start by using AI features already in your software, then pilot one or two high-volume use cases such as customer support, data entry or lead qualification, measure the results, and scale what works.

Is it better to build custom AI or use AI built into existing software?

Use embedded AI for productivity inside a single tool, buy specialist products for common tasks, and build custom AI where it connects several systems, uses your proprietary data or shapes customer experiences that set you apart. Many businesses combine all three.

How much does it cost to implement AI in a business?

Costs depend on the use case, data preparation, integrations, model usage and ongoing support, so there is no single price. Plan for a one-time build and integration cost plus recurring running costs, and start with a focused pilot to validate the business case before scaling.

How can CodeBase Coders help us turn AI spending into results?

CodeBase Coders starts with a short discovery to find the AI use cases with the clearest payback, then builds and integrates them into the systems you already run: AI chatbots and assistants, workflow automation, document and data processing, and custom AI features inside your web or mobile apps. We connect AI to your CRM, ERP, WhatsApp and other tools through secure APIs, keep humans in the approval loop, and support everything after launch. Explore our AI solutions, software integration services and technology consulting, or book a free consultation.

Sources

• Gartner: Gartner Forecasts Worldwide AI Spending to Grow 49.5% in 2026 (16 September 2026)
• CIO.com: AI spend will jump 49.5% in 2026, says Gartner (23 September 2026)
• TechEdge AI: Gartner Forecasts $2.7 Trillion in AI Spending in 2026 (segment table)
• Gartner: Gartner Forecasts Worldwide AI Spending to Grow 47% in 2026 (19 May 2026)
• Gartner: Worldwide AI Spending Will Total $2.5 Trillion in 2026 (15 January 2026)
• Gartner: Worldwide IT Spending to Grow 14.2% in 2026, Totaling $6.37 Trillion (27 July 2026)

Work With CodeBase Coders

AI budgets are growing fast, but results come from choosing the right use cases and integrating them properly. CodeBase Coders designs, builds and integrates AI into the software and workflows your business already runs, from chatbots and WhatsApp automation to custom AI features, data pipelines and CRM and ERP integrations.

• AI Development & Integration
• Software & API Integration Services
• Custom Software Development
• Technology & AI Consulting
• Dedicated Development Teams

Ready to move? Book a free consultation with CodeBase Coders and our engineers will map the right AI use cases, team and roadmap to your goals, with no obligation. Explore everything we build at codebasecoders.com.