Deep dive: The state of AI adoption in 2026

In 2026, AI's move from pilot programs to production affected enterprise workforces in unanticipated ways. All the while, profitability gains from AI remained stagnant across sectors. Organizations that feel like they're lagging might find they're in good company with most of their peers.

The findings in this deep dive illustrate where AI adoption stood in 2025, developments in 2026 and the realities we've seen so far. Some of the most interesting findings include the scaling bottleneck some organizations are experiencing with AI technology, particularly agentic AI, and the EBIT plateau, where only about a third of organizations report any positive financial return attributed to AI so far.

Here are some figures from 2025 and 2026 on AI adoption that can help business leaders see how progress has unfolded and where they might need to tweak their roadmap for adoption and implementation in 2027.

Current AI use and adoption

The U.S. Census Bureau's Business Trends and Outlook Survey (BTOS) measured the percentage of U.S. firms actively using AI in their business operations. This information reflects data collected between December 2025 and May 2026, and shows how large enterprises are leading the charge in active AI use among their employees:

Diving deeper:

• The BTOS survey also reported that 20% to 23% of firms plan to use AI by the end of 2026.
• According to the BTOS survey, the information, finance and insurance sectors led the way in U.S. enterprise AI use.
• McKinsey & Company's "State of AI in 2026: On the road to ROI" report found that 89% of the 1,719 participants surveyed globally used AI in at least one business function.

How companies are scaling AI deployments

AI is moving beyond experimentation, with most companies either piloting or scaling their deployments in 2026. McKinsey's "State of AI" survey found that scaling AI is becoming the standard:

Diving deeper:

• McKinsey found that large enterprises led the way in scaling, aligning its report with the Census Bureau's findings of active AI use. Just over 50% of large enterprises -- those with at least $1 billion in revenue -- reported scaling AI across their organization.
• Those who aren't scaling aren't too far behind. In Deloitte's report, "State of AI in the Enterprise: The untapped edge," the authors found that, in a survey of 3,235 business leaders globally, only 25% had moved 40% or more of their AI experiments into production.
• In a 2026 Gartner survey of 1,303 enterprise respondents whose enterprise had annual revenue of at least $50 million, only 22% reported successfully scaling AI across multiple business units.

Agentic AI

AI agents caught the C-suite's attention in 2026. In its "State of AI" report, Deloitte found that 74% of companies plan to deploy agentic AI within two years. PwC's "AI Agent Survey" shared what 300 senior executives found were the top five most popular business use cases for agentic AI:

Diving deeper:

• IBM's "Rewiring the C-suite: The fast track to 2030" report found that 72% of CEOs predicted agentic AI would be a primary growth driver by 2030.
• The most popular agentic AI use cases in PwC's survey closely matched the most popular overall AI use cases from Gartner's 2026 survey. The Gartner survey found the top three most popular use cases were cybersecurity threat detection and response, IT service desk automation and automated code generation.
• According to Deloitte's report, only 21% of companies have a mature governance model for autonomous agents.

Governance, risk and sovereignty

AI is still a new technology that businesses are struggling to manage effectively and securely. Risks associated with AI deployments are top of mind for executives, and Deloitte's "State of AI" report shared the five most pressing risks for C-suite executives:

Diving deeper:

• Gartner's "Quarterly Emerging Risk Report" for Q3 2026 shared 308 enterprise risk management leaders' top concerns, which included AI discovery of cyber vulnerabilities, AI strategy miscalculation and AI-driven competitive displacement.
• McKinsey's "State of AI" report showed that, perhaps unsurprisingly, enterprises identified as high performers are more likely to mitigate AI-driven exploitation of vulnerabilities.
• Both Deloitte and IBM found that 83% of business leaders consider AI sovereignty to be at least moderately important, if not essential, to their AI strategies.

ROI from AI

Many businesses expected AI to deliver positive ROI in 2026. Deloitte's "State of AI" report found that while 74% of organizations hoped to grow revenue through AI, only 20% achieved this goal. In its report, "From tokenmaxxing to ROI: A reality check for enterprise AI," Google Cloud, in association with the National Research Group, surveyed 2,403 executives globally and found a similar result: only 26% of enterprises reported accelerating year-over-year returns. These are what Google dubbed the "AI ROI leaders." Here's where all enterprises fell according to Google Cloud's research:

Diving deeper:

• McKinsey's "State of AI" report found that 37% of survey respondents attributed some EBIT impact to AI; however, this figure had plateaued starting back in 2025.
• According to IBM's "2026 CEO Study," nearly 80% of executives expect AI to drive significant revenue by 2030, but only 24% know where it's going to come from.
• Deloitte's "State of AI" report also shared that 84% of organizations are increasing their AI investments, and 78% of leaders reported greater confidence in AI.

Key takeaways and predictions for 2027

Large enterprises are leading the way in AI adoption, scaling and ROI, and the main sectors seeing benefits from AI are information, finance and insurance, all of which benefit from the automation AI provides.

---

AI has found its foothold within enterprises, but without specialized tooling, its progress will continue to plateau.

---

However, the data doesn't show a significant gap between large enterprise adoption and other organizational adoption. Only about a quarter of organizations have moved 40% or more of their AI experiments into production, and only 22% of organizations have scaled AI across multiple business units. While most organizations can attribute some ROI to AI, EBIT figures are plateauing across organizations, and nearly 20% of organizations are seeing inconsistent, plateauing or declining AI ROI.

Here's the prediction for enterprise AI in 2027: AI has found its foothold within enterprises, but without specialized tooling, its progress will continue to plateau. Enterprises that have found use cases for AI automation will continue to use the technology as they already do and might discover a couple of new use cases through continued experimentation. But until AI and agentic AI become more specialized tools, their deployments will only become more expensive for work that is largely comparable to what people are already capable of. For now, most organizations shouldn't feel like they are falling behind the AI curve. But as we head into the new year, businesses must look closely at their AI investments and ensure they are implementing technology that will be more of a benefit than a burden.

Everett Bishop is an associate site editor for TechTarget's AI & Emerging Tech group, covering AI, quantum computing and other emerging technologies.