The Fifth Annual Domino Enterprise AI Report
Domino Data Lab's Fifth Annual Domino Enterprise AI Report (July 2026; n=639 senior AI leaders, executive survey) found that 57% of enterprises say AI returns are not keeping pace with investment, unchanged since 2025, even as 93% report improved ability to move AI into production.
Key findings
- 0157% of enterprises say AI ROI fails to outpace their investment, the same share as in 2025.Over the same period, the share reporting improved production capability rose from 88% to 93%.
- 02The share reporting ROI at or below investment was 51.1% in North America, 66.9% in the UK and 67.0% in continental Europe.North American firms reported better returns but were also more likely to give business users no direct access to AI insights (12.8%).
- 0343% have agentic AI running in governed production, while 41% are piloting (12%) or scaling (29%) agents without governance.Expanding agentic AI tied with upskilling business users as the top 2026 priority, each at 38.5%.
- 0440% still rely entirely on at least one mediated access method, such as scheduled reports or analyst requests, to get AI insights to business users.Domino calls this a 'last-mile gap' between models in production and the people who act on them.
- 0575% of organizations with fully integrated AI governance report significantly improved AI delivery speed, versus 23% where governance is falling behind.This is a correlation from the survey, not evidence that governance causes faster delivery.
By the numbers
93%report improved ability to move AI into production
41%piloting or scaling agentic AI without governance
40%rely on mediated access to AI insights
What it means for you Draft
For executives at $10–100M companies
Even large, regulated enterprises with dedicated AI teams report that returns have not caught up with spending, two years running. One lesson that may carry over to a $10–100M company is that getting a model working is not the same as getting its output into the hands of the people who make decisions. Before adding agents, it may be worth deciding who is accountable for what they do.
For practitioners
Plan how business users will actually see and act on AI output, not just how the model will be deployed. Put basic agent governance, such as logging, approvals and ownership, in place before scaling agents.
Limitations
Medium trust.Disclosed sample and independent fieldwork by BARC, but self-reported ROI and a vendor that sells AI governance software.
Respondents are AI leaders at $100M+ regulated enterprises, so results do not map directly to smaller firms. ROI is self-reported, and Domino sells the governance and deployment platform its conclusions favor; the full report is gated.
About the publisher: Sells an enterprise AI platform for building, deploying and governing models and agents.
Domino Data Lab. "The Fifth Annual Domino Enterprise AI Report." July 21, 2026. https://domino.ai/enterprise-ai-report
Related reports
Global AI Pulse Q3 2026: AI at scale: Accountability, resilience and economics34% report significant employee adoption of AI agents, up from 25% in Q1 2026.
The State of AI in the Enterprise 2026: The untapped edge34% are using AI to deeply transform the business; 30% are redesigning key processes; 37% use AI at a surface level.
EY AI Risk and Governance Survey 202698% of large US public companies have AI governance policies, but 47% have skipped the process for urgent deployments and 36% reported a materially harmful AI incident.