Agentic AI Statistics (2026)

A data analysis of how many companies actually run AI agents, what those agents do, why so many agent projects stall, and how big the market really is, with every figure traced to its publisher.

Agentic AI is still mostly a plan. Only 17% of organizations have deployed AI agents, per Gartner’s 2026 CIO survey, and Deloitte finds 23% using agentic AI at least moderately. Ambition runs far ahead of execution: more than 60% expect to deploy within two years, yet Gartner also predicts over 40% of agentic AI projects will be canceled by the end of 2027. The numbers below cover adoption, tasks, failure rates, ROI, market size, and trust, plus what they mean when you choose between agent platforms and simpler automation tools.

Key takeaways

• 17% of organizations have deployed AI agents, and more than 60% expect to within two years, per Gartner. Deloitte’s 23% “at least moderately” lands in the same range.
• Over 40% of agentic AI projects will be canceled by the end of 2027, Gartner predicts, due to escalating costs, unclear business value, or inadequate risk controls (Gartner)
• 40% of companies with $1B+ revenue now scale AI agents, up from 27%, while smaller organizations stayed at 22% (McKinsey)
• Only 2% of organizations run AI agents at full scale, with 12% at partial scale and 23% in pilots (Capgemini)
• The AI agents market is sized at $7.84B-$7.92B for 2025, per MarketsandMarkets and Precedence Research. Gartner’s broader count of AI agent software spending hits $206.5B in 2026.
• Trust in fully autonomous agents fell from 43% to 27% in one year (Capgemini), and only 21% of companies have a mature governance model for them (Deloitte)
• 32% of agent builders name quality as their top barrier to production, ahead of latency at 20% (LangChain)
• 32% of organizations skipped a software purchase because coding agents let them build it in-house (McKinsey)
• Salesforce’s Agentforce ARR passed $1.5B, up over 240% year over year, the clearest vendor revenue signal so far (Salesforce)

How many organizations use AI agents in 2026?

Between 17% and 23% of organizations have AI agents in meaningful use, depending on how a survey defines it. Gartner puts deployment at 17%, Deloitte finds 23% using agentic AI at least moderately, and McKinsey found 23% scaling agents somewhere in 2025. Looser definitions push the figure past 60%.

1. Only 17% of organizations have deployed AI agents, but more than 60% expect to within two years, per Gartner’s Hype Cycle, citing its 2026 CIO and Technology Executive Survey. Gartner calls it the most aggressive adoption curve of any emerging technology it measured.
2. 23% of companies use agentic AI at least moderately today, and 74% expect to within two years, per Deloitte’s State of AI survey of 3,235 leaders, fielded in August and September 2025. Deloitte projects 23% will use it extensively and 5% will fully integrate it by then.
3. 23% of organizations were scaling an agentic AI system in 2025, and another 39% were experimenting, per McKinsey’s 2025 survey of 1,993 participants. “Scaling” meant expanding deployment within at least one business function, which is a low bar.
4. No more than 10% of organizations were scaling agents in any single business function, in the same McKinsey survey. Agents are spreading wide but shallow.
5. 62% of large U.S. organizations are building, deploying, or developing AI agents, up from 53% one quarter earlier, per the KPMG AI Pulse survey of 314 leaders at $1B+ companies (July-August 2026). This counts development work, not only live agents.
6. 79% of senior executives say AI agents are already being adopted in their companies, per PwC’s AI Agent Survey of 300 executives in May 2025. Yet 68% say half or fewer of their employees interact with agents in everyday work.
7. 61% of CEOs say they are actively adopting AI agents and preparing to implement them at scale, per an IBM study of 2,000 CEOs published in May 2025.
8. 57.3% of agent builders have agents in production, and another 30.4% are developing them with concrete plans to deploy, per LangChain’s State of Agent Engineering survey of 1,340 respondents (November-December 2025). The sample skews technical: 63% work in technology.
9. 81% of leaders expect agents to be moderately or extensively integrated into their AI strategy within 12-18 months, per Microsoft’s 2025 Work Trend Index, which drew on survey data from 31,000 workers in 31 countries.

Sources: Gartner, Deloitte, McKinsey, Capgemini, LangChain, KPMG, and PwC, linked above. Shares are not directly comparable because each survey defines adoption differently.

The honest answer is about one in five, if “use” means agents doing real work. The 60-79% figures measure intent, pilots, and development. For AI use beyond agents, see our AI adoption statistics.

How many companies are piloting vs. scaling AI agents?

Most agent programs are still pilots or experiments. McKinsey’s 2026 survey shows 40% of companies with $1 billion or more in revenue scaling agents in at least one function, up from 27% a year earlier. Smaller organizations held at 22%. Capgemini found just 2% running agents at full scale.

10. 40% of large organizations now scale AI agents, up from 27% in 2025, while smaller organizations stayed at 22%, per McKinsey’s State of AI 2026, fielded May-June 2026 with 1,719 participants. The gap between big and small companies is widening.
11. Only 2% of organizations have deployed AI agents at scale, 12% at partial scale, 23% have launched pilots, and 61% are exploring, per the Capgemini Research Institute survey of 1,500 executives in 14 countries (July 2025).
12. Agent deployment nearly quadrupled in six months, to 42% of organizations with at least some agents deployed, up from 11%, per KPMG’s Q3 2025 Pulse. KPMG’s broader 2026 measure, which also counts development work, reached 62%.
13. 25% of large organizations are developing or implementing multi-agent systems, up from 6% in each of the previous two quarters, per KPMG (September 2026). Multi-agent systems coordinate several specialized agents on one workflow.
14. In a January 2025 Gartner poll of 3,412 webinar attendees, 19% had made significant investments in agentic AI, 42% conservative investments, 8% none, and 31% were waiting or unsure (Gartner).
15. 60% of organizations evaluated enterprise-grade AI tools, 20% reached the pilot stage, and 5% reached production, per the MIT NANDA report “The GenAI Divide” (July 2025). It covers custom and vendor GenAI systems broadly, not agents alone.
16. Of companies adopting agents, 35% are doing so broadly and 17% say agents are fully adopted in almost all workflows and functions, per PwC.

Source: McKinsey, The State of AI in 2026, Exhibit 2. 2026 samples: 1,035 smaller and 595 larger organizations.

What tasks do AI agents handle?

Agents scale first in technical back-office work. McKinsey finds 12% of organizations scaling agents in IT, 11% in knowledge management, and 10% in software engineering, against 3% in manufacturing. Among builders surveyed by LangChain, customer service (26.5%) and research and data analysis (24.4%) are the leading use cases.

17. IT (12%), knowledge management (11%), and software engineering (10%) lead scaled agent use, followed by service operations and product development (9% each) and marketing and sales (8%), per McKinsey. Manufacturing trails at 3%.
18. 31% of technology-industry respondents report scaling agents in software engineering, the highest cell in McKinsey’s industry breakdown.
19. About two in ten organizations are scaling software coding agents, rising to 31% at larger enterprises (McKinsey). At larger enterprises, that tops the 25% scaling other types of AI agents.
20. Customer service is the most common primary agent use case at 26.5%, with research and data analysis at 24.4%, per LangChain. Together they make up more than half of primary deployments.
21. More than half of companies use or plan to use agents within six months in customer service (57%), sales and marketing (54%), and IT and cybersecurity (53%), per PwC.
22. Technology departments lead agent deployment, with 95% now using agents, per KPMG’s Q3 2025 Pulse.
23. 46% of leaders say their companies use agents to fully automate workflows or processes, per Microsoft (April 2025).
24. 85% of companies expect to customize agents to fit their business, and leaders expect the highest agentic impact in customer support, per Deloitte.

Service teams are the furthest along on customer-facing agents. Our customer support software statistics track how service organizations deploy them.

Why do agentic AI projects fail or get canceled?

Gartner predicts over 40% of agentic AI projects will be canceled by the end of 2027 due to escalating costs, unclear business value, or inadequate risk controls. Teams building agents name quality as the top barrier: 32% cite it, per LangChain, ahead of latency at 20%.

25. Over 40% of agentic AI projects will be canceled by the end of 2027, Gartner predicted in June 2025 (Gartner). Its analyst described most current projects as “early stage experiments or proof of concepts that are mostly driven by hype.”
26. Only about 130 of the thousands of agentic AI vendors are real, by Gartner’s estimate (Gartner). The rest practice “agent washing,” rebranding assistants, RPA, and chatbots as agents.
27. By 2027, 40% of enterprises will demote or decommission autonomous AI agents because of governance gaps found only after production incidents, per a May 2026 Gartner release.
28. 32% of agent builders cite quality as their top barrier to production, and 20% cite latency (LangChain). Quality covers accuracy, consistency, and staying within policy.
29. Cybersecurity and cost tie as the top-ranked agent challenges at 34% each, per PwC, while connecting agents across applications (19%), organizational change (17%), and employee adoption (14%) rank lowest. PwC argues those bottom three are the real blockers.
30. About 20% of organizations say AI operating costs, including tokens, constrained their AI use, and about one in ten say costs constrained AI agents specifically (McKinsey).
31. 95% of organizations are getting zero return from $30-40 billion in enterprise GenAI investment, per MIT NANDA (July 2025). The finding covers GenAI broadly and rests on 153 survey responses, 52 interviews, and over 300 public initiatives, so treat it as directional.
32. Only 25% of AI initiatives have delivered expected ROI, and only 16% have scaled enterprise wide, CEOs told IBM in 2025.

The most useful line in Gartner’s cancellation release is its advice: start “by using AI agents when decisions are needed, automation for routine workflows and assistants for simple retrieval.” Its analyst added that “many use cases positioned as agentic today don’t require agentic implementations.” For rule-based work, a deterministic workflow in Zapier, Make, or n8n is easier to test, predict, and debug than an agent.

Do AI agents deliver ROI?

Adopters report productivity gains first and profit later. In PwC’s survey, 66% of companies adopting agents report increased productivity and 57% report cost savings. Enterprise-wide impact is rarer: McKinsey finds 37% of organizations attribute any EBIT impact to AI, about the same share as a year earlier.

33. 66% of agent adopters report increased productivity, 57% cost savings, 55% faster decision-making, and 54% improved customer experience (PwC). PwC notes that early value tends to come from internal use cases.
34. 58% of large organizations report measurable business value from AI (KPMG report), led by productivity (55%), faster decisions (49%), and stronger financial performance (37%), per KPMG’s release (September 2026). This covers AI broadly, with agents a growing share.
35. 37% of organizations attribute at least some EBIT impact to AI, and AI high performers hold flat at about 6% of respondents (McKinsey).
36. High performers are 2.7 times more likely than others to scale AI agents beyond coding, and nearly three-quarters of them have fundamentally redesigned workflows, against one-quarter of other respondents (McKinsey).
37. About 80% of organizations piloting or deploying autonomous capabilities report workforce reductions, but cut rates were nearly equal between higher-ROI adopters and those with modest or negative results, per a Gartner survey of 350 executives (Q3 2025). Layoffs free budget; they do not create return.
38. 57% of leaders expected measurable ROI from agents within 12 months, per KPMG’s Q3 2025 Pulse.
39. 85% of CEOs expect their scaled AI efficiency and cost-saving investments to return a positive ROI by 2027 (IBM).

The ROI is real but concentrated. It shows up where companies redesign the workflow around the agent, which is a process decision before it is a software decision.

How big is the AI agent market?

Estimates span a wide range because definitions differ. Narrow “AI agents market” reports size 2025 at $7.84 billion (MarketsandMarkets) to $7.92 billion (Precedence Research). Gartner’s broader count of AI agent software spending, which includes agents embedded in existing applications, reaches $206.5 billion in 2026 (Gartner).

40. The AI agents market will grow from $7.84 billion in 2025 to $52.62 billion by 2030, a 46.3% CAGR, per MarketsandMarkets (April 2025).
41. Precedence Research sizes the market at $7.92 billion in 2025 and $11.55 billion in 2026, reaching $294.66 billion by 2035 at a 43.57% CAGR (Precedence). The two firms nearly match on the 2025 base; their long-range forecasts diverge.
42. Spending on AI agents and assistants will reach $29.2 billion in 2026, up from $16.5 billion in 2025, and $65.5 billion in 2027, per Gartner’s September 2026 forecast. That sits inside total AI spending of $2.7 trillion in 2026, up 49.5%.
43. AI agent software spending, a broader Gartner category, reaches $206.5 billion in 2026 and $376.3 billion in 2027, up from $86.4 billion in 2025 (Gartner).
44. In Gartner’s best case, agentic AI drives about 30% of enterprise application software revenue by 2035, surpassing $450 billion, up from 2% in 2025 (Gartner).
45. IDC forecasts AI spending, driven by agentic AI, will reach $1.3 trillion in 2029 and exceed 26% of worldwide IT spending, growing 31.9% a year between 2025 and 2029 (IDC).
46. AI agents could generate up to $450 billion in economic value by 2028 across the markets Capgemini surveyed (Capgemini).
47. Salesforce’s Agentforce ARR exceeded $1.5 billion, up over 240% year over year, in its fiscal Q2 2027 results (Salesforce). The figure now includes Salesforce’s AI offerings, Slackbot, and Headless 360, and the company delivered 3.2 billion “Agentic Work Units” in the quarter, up 97% quarter over quarter.

Sources: MarketsandMarkets (April 2025), Precedence Research, and Gartner forecasts (May and September 2026), linked above. Gartner’s agents-and-assistants figures are converted from millions and rounded.

How much are companies budgeting for AI agents?

48. 88% of executives plan to raise AI budgets in the next 12 months because of agentic AI, and over a quarter plan increases of 26% or more (PwC).
49. Large U.S. organizations project $173 million in AI investment over the next 12 months, up from $130 million a year earlier (KPMG Q3 2026).
50. 28% of organizations spend more than 10% of their ICT budget on AI, and 60% expect to increase AI investment next year (McKinsey).

Do companies trust AI agents?

Trust is falling as agents gain autonomy. Capgemini found trust in fully autonomous AI agents dropped from 43% to 27% in one year. Deloitte reports only 21% of companies have a mature governance model for autonomous agents, even as 74% expect to use agentic AI at least moderately within two years.

51. Trust in fully autonomous AI agents fell from 43% to 27% in one year (Capgemini).
52. Only 21% of companies report a mature governance model for autonomous agents (Deloitte). Deloitte warns that rushing to deploy before governance is in place exposes companies to significant risk.
53. Executives trust agents most for data analysis (38%) and performance improvement (35%), and least for autonomous employee interactions (22%) and financial transactions (20%), per PwC. 28% rank lack of trust as a top-three challenge.
54. 49% of large organizations have defined high-risk use cases where agents are not allowed to make autonomous decisions (KPMG). 74% now include cost reviews in AI approvals, up from 61%; 70% use monitoring dashboards; and 43% enforce usage or token budgets.
55. Confidence in AI governance rose to 73% from 57% in one quarter, yet the share building controls into agents fell to 30% from 43% two quarters earlier (KPMG). Confidence is rising faster than controls.
56. 89% of agent builders have implemented observability, but only 52% run evaluations (LangChain).
57. At least 15% of day-to-day work decisions will be made autonomously through agentic AI by 2028, up from 0% in 2024, Gartner predicts (Gartner).

Are AI agents replacing software purchases?

They are starting to. McKinsey reports 32% of organizations decided against buying at least one software product or feature because agentic coding tools let them build it in-house. Gartner estimates up to $234 billion of enterprise application spending is exposed to “agentic arbitrage” through 2030.

58. 32% of organizations skipped at least one software purchase in favor of building with agentic coding tools, led by technology (41%) and healthcare (39%) respondents (McKinsey). Nearly half of AI high performers report this, against 31% of others.
59. Up to $234 billion of enterprise application spending is exposed to agentic arbitrage through 2030, roughly 20% of enterprise application SaaS spending, per Gartner (July 2026). When agents complete tasks across systems, fewer people log into each app, which breaks seat-based pricing.
60. By 2028, over half of enterprises will stop paying for assistive AI such as copilots and favor platforms that commit to workflow results (Gartner). Vendors that bolt AI onto legacy apps face margin compression of up to 80% by 2030.
61. 40% of enterprise applications will integrate task-specific AI agents by the end of 2026, up from less than 5% in 2025 (Gartner). Gartner separately expects 33% of enterprise software applications to include agentic AI by 2028, up from less than 1% in 2024 (Gartner).
62. External partnerships reached deployment about 67% of the time, against about 33% for internally built tools, in MIT NANDA’s sample, which the authors note is self-reported.

The numbers point two ways: coding agents make building cheaper, while MIT’s data says bought tools reach production more often. For most teams, agents will arrive inside software they already pay for. Our no-code automation statistics show how far simpler workflow tools already reach, and our guide for CTOs covers build-versus-buy picks.

What are the agentic AI predictions for 2027-2030?

Analysts expect agents to move from single tasks to coordinated teams. Gartner predicts one-third of agentic AI implementations will combine agents with different skills by 2027, a third of user experiences will shift to agentic front ends by 2028, and at least 50% of knowledge workers will build agent skills by 2029.

63. By 2027, one-third of agentic AI implementations will combine agents with different skills to manage complex tasks (Gartner).
64. By 2028, a third of user experiences will shift from native applications to agentic front ends, per the same Gartner forecast.
65. By 2029, at least 50% of knowledge workers will develop new skills to work with, govern, or create AI agents (Gartner).
66. 38% of organizations will have AI agents as team members within human teams by 2028, and 15% of business processes are expected to reach semi- or full autonomy within 12 months (Capgemini).
67. 32% of managers plan to hire AI agent specialists within 12-18 months, and 28% are considering hiring AI workforce managers (Microsoft).
68. Gartner predicts autonomous business will be a net-positive job creator by 2028 to 2029, driven by new forms of work AI cannot absorb (Gartner).

What to do with this data

• Start with the workflow, not the agent. Gartner says many use cases sold as agentic do not need agents. If the steps are fixed and rule-based, a deterministic workflow from our automation category is cheaper to govern. Save agents for steps that need judgment.
• Test for agent washing. Gartner counts about 130 genuine agentic vendors among thousands. Ask each vendor which steps the agent plans and executes on its own, and what happens when it is wrong.
• Budget for running costs, not just licenses. About 20% of organizations say AI operating costs constrained their use (McKinsey), and 43% of large firms now enforce token budgets (KPMG).
• Put governance in place before scale. Only 21% of companies have a mature agent governance model, and Gartner expects 40% of enterprises to demote or decommission agents after incidents. Define which decisions an agent can never make alone.
• Prefer agents inside tools you already own. With 40% of enterprise apps expected to embed task-specific agents by the end of 2026, the agent you need is often a feature, not a new platform. Fewer, better-adopted tools still win. See how we score tools on our about page.
• Read the adjacent data. Our ChatGPT statistics, no-code automation statistics, and AI adoption statistics cover the tools most agent projects start from.

Frequently asked questions

What percentage of companies use AI agents?

About one in five use agents for real work. Gartner reports 17% of organizations have deployed AI agents, and Deloitte finds 23% using agentic AI at least moderately. Broader measures run higher: 62% of large U.S. firms are building, deploying, or developing agents, per KPMG.

What percentage of agentic AI projects fail?

Gartner predicts over 40% of agentic AI projects will be canceled by the end of 2027, citing escalating costs, unclear business value, and inadequate risk controls. A separate Gartner forecast expects 40% of enterprises to demote or decommission autonomous agents by 2027 after governance gaps surface in production.

How big is the AI agent market?

Narrow estimates put the AI agents market at $7.84 billion (MarketsandMarkets) to $7.92 billion (Precedence Research) in 2025, with MarketsandMarkets projecting $52.62 billion by 2030. Gartner’s broader AI agent software spending figure, which counts agents embedded in applications, reaches $206.5 billion in 2026.

What are AI agents used for most?

IT, knowledge management, and software engineering lead scaled use, at 12%, 11%, and 10% of organizations, per McKinsey. Among agent builders, customer service (26.5%) and research and data analysis (24.4%) are the top primary use cases, per LangChain. Manufacturing trails at 3%.

Do AI agents deliver ROI?

Most adopters see productivity gains: 66% in PwC’s survey, with 57% reporting cost savings. Enterprise-level profit impact is rarer. McKinsey finds 37% of organizations attribute any EBIT impact to AI, and only about 6% qualify as high performers, unchanged from 2025.

Do businesses trust AI agents?

Less than they did. Capgemini found trust in fully autonomous agents fell from 43% to 27% in a year. Only 21% of companies have a mature governance model for autonomous agents, per Deloitte, and just 20% of executives trust agents with financial transactions, per PwC.

What is the difference between AI agents and automation tools?

Automation tools follow fixed rules; agents plan and execute multi-step work on their own. Gartner advises using agents when decisions are needed, automation for routine workflows, and assistants for simple retrieval. It also estimates only about 130 of the thousands of agentic AI vendors are real.

Will AI agents replace SaaS tools?

Partly. Gartner estimates up to $234 billion of enterprise application spending, about 20% of SaaS application spend, is exposed to agents through 2030. Already, 32% of organizations have skipped a software purchase because coding agents let them build it, per McKinsey.

How we compiled these statistics

We drew on 24 pages from 14 publishers: research and advisory firms, consultancies, a university lab report, a developer-tools survey, and one vendor earnings release. Every figure was retrieved from the publisher’s page or PDF in October 2026; where a publisher blocked automated access, we read the Internet Archive’s copy of the same page. Figures we could not trace to a primary source were omitted, and where estimates disagree we show the range and cite each source.

Sources

• Gartner: 2026 Hype Cycle for Agentic AI (2026)
• Gartner: Over 40% of Agentic AI Projects Will Be Canceled by End of 2027 (2025)
• Gartner: 40% of Enterprise Apps Will Feature Task-Specific AI Agents by 2026 (2025)
• Gartner: Uniform Governance Across AI Agents Will Lead to Enterprise AI Agent Failure (2026)
• Gartner: Autonomous Business and AI Layoffs Do Not Deliver Returns (2026)
• Gartner: $234 Billion in Enterprise Application Software Spend at Risk from Agentic AI (2026)
• Gartner: Most Enterprises to Abandon Assistive AI for Outcome-Focused Workflow by 2028 (2026)
• Gartner: Worldwide AI Spending to Grow 49.5% in 2026 (2026)
• McKinsey: The State of AI in 2026, On the Road to ROI (2026)
• McKinsey: The State of AI in 2025, Agents, Innovation, and Transformation (2025, archived)
• Deloitte: State of AI in the Enterprise, The Untapped Edge (2026)
• PwC: AI Agent Survey (2025)
• KPMG: AI Quarterly Pulse Survey Q3 2026 news release (2026)
• KPMG: AI Quarterly Pulse Survey Q3 2026 report (2026)
• KPMG: Agent Deployment Accelerates, Q3 2025 AI Pulse (2025)
• Capgemini Research Institute: Rise of Agentic AI (2025)
• LangChain: State of Agent Engineering (2026)
• IBM: CEO Study, CEOs Double Down on AI (2025)
• Microsoft: 2025 Work Trend Index, The Year the Frontier Firm Is Born (2025)
• MIT NANDA: The GenAI Divide, State of AI in Business 2025 (2025, archived)
• MarketsandMarkets: AI Agents Market Worth $52.62 Billion by 2030 (2025)
• Precedence Research: AI Agents Market Size (2026)
• IDC: Agentic AI to Dominate IT Budget Expansion (2025)
• Salesforce: Fiscal 2027 Second Quarter Results (2026)