Download PDF

The State of Enterprise AI Adoption

Research report of 600+ business and IT leaders

Download PDF Scroll to explore

Table of Contents

  1. Foreword
  2. Executive Summary
  3. 01. AI has moved from experimentation to enterprise-scale execution
  4. 02. Scaling AI means rethinking how people and agents will work together
  5. 03. Deeper AI adoption is linked to stronger revenue outcomes
  6. 04. What enterprise AI Adoption Leaders do differently
  7. 05. How organizations are approaching AI governance
  8. 06. The challenges slowing AI adoption
  9. 07. How organizations measure AI value and ROI
  10. Conclusion: The next phase of enterprise AI adoption
  11. About the Survey
  12. Appendix
  13. About Creatio
Foreword

Every business will compete on how well its people and AI agents work together

AI is here and now, and it has real traction in many business initiatives. Our survey of 608 IT and business decision-makers shows that nearly all organizations have strong AI ambitions, and more than two-thirds have deployed, are scaling, or have AI embedded across the organization.

Achieving the right mix of people, workflows, applications, and AI agents will be important for every organization. And achieving that combination will drive AI technology decisions to the forefront of competitive and go-to-market strategy.

On behalf of everyone at Creatio, I hope the insights in this report help your organization to adopt AI in a transformational way and propel your operations into the future with unlimited potential for growth.

John Arnold
John Arnold
Head of Product Marketing and Strategic Advisory, Creatio
78%
of organizations are scaling AI agents within twelve months
Don’t wait for the future of AI

If your AI plan still runs on pilots, you are a year behind the market. Set a new plan to scale AI agents across all departments.

Executive Summary

Seven findings shaping the next phase of enterprise AI

Enterprise AI adoption is becoming an operating model decision. The opportunity is to embed AI in the workflows that generate revenue, serve customers, and coordinate work across the business.

Doing this well requires executives to decide how people and agents share responsibilities, how decisions are governed, and how success is measured.

Creatio’s survey of 608 IT and business decision-makers shows that AI is already delivering business benefits. It also reveals a gap between organizations using AI and those adopting it broadly enough to qualify as AI Adoption Leaders. Leaders represent 17% of this sample. Their experience offers a useful benchmark for the other 83%.

1. AI moves beyond experimentation — 70% are already putting it to work

Seven in ten organizations are already beyond AI experimentation. 31% have deployed AI in selected functions, while another 39% are scaling it across the organization or have embedded it enterprise-wide.

The next wave is already forming: more than three quarters expect to scale AI agents across departments or use cases within twelve months.

The takeaway

AI is moving from isolated pilots into the operating fabric of the enterprise. The management challenge is no longer whether to adopt AI, but how to coordinate and scale it across the business.

70% are past experimentation
  • Still evaluating 13%
  • Experimenting 15%
  • Deployed in selected functions 31%
  • Scaling across the organization 25%
  • Embedded across the organization 14%

2. The AI workforce revolution: redefining work, not replacing people

Only 9% expect significant headcount reduction from AI. 89% of respondents expect AI agents to expand workforce capacity, augment existing teams, or create new roles.

Specifically:

  • 38% expect greater capacity and growth for current staff.
  • 34% expect augmentation without major headcount change.
  • 17% expect new roles.
  • 9% expect job cuts.
  • 2% are unsure.

Nearly twice as many respondents expect AI agents to create new roles as expect them to significantly reduce headcount—17% versus 9%.

The takeaway

Organizations need to plan how people and AI agents will work together: which tasks agents handle, where human judgment matters, and who owns the outcomes.

89%
expect AI agents to augment people and create new roles
9%
Expect headcount reduction
  • 38% Create capacity and growth for current staff
  • 34% Augment teams without major headcount change
  • 17% Create new roles
  • 9% Reduce headcount
  • 2% Unsure

3. Companies scaling AI are 2x more likely to report revenue gains as a positive impact of AI

When asked which positive outcomes they had experienced from AI in the past six months, organizations with deeper adoption were more likely to select revenue gains. Among organizations automating more than half of their workflows with AI, 33% report increased revenue from AI in the past six months, compared with 14% among those automating less than 10%.

This suggests that as AI becomes more deeply embedded in workflows, revenue gains are more commonly reported among its benefits.

The takeaway

This places enterprise AI adoption firmly on the growth agenda. For executives, the implication is to look beyond productivity and efficiency alone and measure its contribution to revenue as adoption expands.

Reported revenue gains
  • Under 10% workflows with AI 14%
  • Over 50% workflows with AI 33%

4. What AI Adoption Leaders do differently: scaling AI enterprise-wide and measuring impact beyond efficiency

Just 17% of surveyed organizations qualify as AI Adoption Leaders, based on the breadth of their AI capabilities and agent adoption across business functions.

The takeaway

Enterprise AI leadership requires adoption depth, broader measures of business value, and the governance to support scale.

  • Automate more workflows with AI: 48% automate over half their workflows with AI, versus 14%.

  • Scale agents across departments: 53% of Leaders, versus 21% of others.

  • Embed AI deeply into business operations: 46% of Leaders report organization-wide integration, versus 8% of others.

  • Measure success more broadly: 93% of Leaders use at least three measures of AI success, versus 76% of others, with greater emphasis on revenue, employee satisfaction, and innovation, not just time savings.

  • Establish stronger governance: 59% of Leaders have fully established AI governance, versus 23% of others.

5. Companies with established governance are 4x more likely to feel ready to scale AI

AI adoption is accelerating, but enterprise governance has yet to catch up. 78% of organizations are already scaling AI agents across departments or use cases, or expect to within a year. Yet only 29% report fully established AI governance.

The data reveal why closing this gap matters. Organizations with established governance are more than four times as likely to feel very confident about scaling AI (71% vs. 17% among those still developing a framework). AI Adoption Leaders are also significantly ahead: 59% have fully established governance, compared with approximately 23% of other organizations.

The takeaway

Build governance alongside AI deployment. Clear decision rights, permissions, human oversight, and visibility into agent actions provide a foundation for confident expansion.

Very confident about scaling AI
71%
Established governance
17%
Developing governance framework

6. The barriers to scaling AI differ by adoption stage

The challenges that slow broader AI adoption change as organizations move from early experimentation to enterprise-wide adoption. Among organizations at the earliest stage of AI adoption, security and compliance is already a significant challenge (41%), alongside AI skills (40%) and organizational readiness (34%).

As adoption matures, skills and organizational readiness become less prominent, declining to 27% and 21% among the most advanced adopters. At the same time, enterprise-scale requirements move to the forefront: security and compliance rises to 52%, while technology capabilities (37%), commercial models and budgets (35%), and change management (35%) become increasingly important.

The takeaway

Scaling AI doesn’t eliminate barriers – it changes them. As organizations mature, the challenge shifts from building the skills and readiness to adopt AI toward establishing the technology, security, governance, and operating model required to run it at enterprise scale.

Barriers shift with AI maturity
40% 27%
AI skills
34% 21%
Readiness
13% 34%
Governance

7. The AI ROI conversation is moving beyond time and cost savings

Organizations still lead with efficiency (56%) and productivity (44%) when pursuing AI agents, but the value case broadens as adoption matures. Leaders put more emphasis on productivity, customer experience, revenue, employee outcomes, and innovation – not just cost and time savings.

The measurement model has not fully caught up. Only 56% of organizations pursuing better customer experience measure customer satisfaction; 40% pursuing employee enablement measure employee satisfaction; and just 31% building for the future measure innovation outcomes.

The takeaway

The next phase of AI ROI is about connecting every strategic reason for adoption to a business outcome that can actually be measured.

What executives should prioritize next

Creatio’s perspective is that enterprise AI adoption should be managed as a continuing business capability. Select workflows with clear commercial or customer value. Design how people and agents will execute them together.

Establish governance and outcome measures as part of that design, then expand based on evidence. This is how executives can turn adoption into a repeatable management discipline.

AI has moved from experimentation to enterprise-scale execution

In the span of one year, enterprise AI moved from a board-level discussion to production with 70% of organizations now past AI experimentation.

Enterprise AI has moved from a long-term strategic priority to an immediate focus for execution. In 2025, 86% of C-level leaders expected AI agents to be critical or important to their organizations’ goals within two to three years. Just one year later, 70% of surveyed organizations have moved beyond AI experimentation.

The next stage is already underway. More than three quarters of organizations say they are scaling agents across departments within twelve months.

This chapter examines how organizations are progressing beyond pilots, expanding adoption across functions, and preparing to make AI part of how the enterprise operates.

Most organizations are running AI, not testing it

70% are past experimentation
13% Still evaluating
15% Experimenting
31% Deployed in selected functions
25% Scaling across the organization
14% Embedded across the organization

Most surveyed organizations have deployed AI, though adoption ranges from selected functions to integration across the enterprise.

Base: 608 IT and business decision-makers. 2% have no plans and are not shown.

Enterprise AI has moved past experimentation

Seven in ten surveyed organizations have moved beyond AI experimentation. Nearly a third (31%) have deployed AI in selected functions, while 39% are already scaling it across the organization or have embedded it throughout the business.

With only 2% reporting no plans to adopt AI, the findings show broad commitment to its use. For the majority already beyond experimentation, the priority is expanding deployment and integrating AI more deeply into business operations.

As adoption grows, enterprise leaders need to coordinate implementation across functions, define how people and AI agents work together, and measure business outcomes. These decisions will shape how effectively organizations translate deployment into enterprise-scale execution.

Expert’s Perspective

AI is moving faster than human behavior can adapt. While 78% of leaders plan to scale AI agents within a year, technology adoption is only half the equation. The real competitive advantage won't come from deploying agents faster, but from reimagining how humans and AI evolve together.

Dr. Ja-Naé Duane
Dr. Ja-Naé Duane
Futurist, Best-Selling Author and AI Researcher, Brown University School of Engineering and MIT CISR Academic Fellow

Most organizations will be scaling AI agents within a year

A year ago, 86% of C-level leaders saw AI agents as critical or important to goals two to three years out. The 2026 timeline compresses that expectation by roughly two years.

More than a quarter of organizations (27%) are already scaling AI agents across departments, 23% plan to within six months, and 28% within a year. Together, 78% expect to be scaling agents across the enterprise within twelve months. Only 6% have no timeline.

What it means

Budgets have moved from pilots to programs. Many vendors shipped agents inside CRM, ERP, and service platforms, and organizations must have a strategy to adopt AI effectively companywide. Scaling agents is no longer a stretch goal for organizations. It must be a real plan, and all organizations need to follow through on their plans and execute quickly to keep up with the competition.

When organizations expect to scale AI agents across departments
78%
27% Already scaling
23% Within six months
28% Within a year
17% One to two years
6% No timeline
2025
86%
of C-level leaders said agents critical or important in two-to-three-year goals.
2026
78%
of all decision-makers say they are scaling agents across departments within twelve months.

Percentages may not total 100% due to rounding.

Base: 608 business and technology leaders. 2025: Creatio Global AI and No-Code Adoption Survey, C-level respondents.

The race to adopt AI spans IT and the front office

74%
Exploring or deploying AI
IT and technical
IT operations, software development, R&D
Unified AI strategy and execution required

Two AI strategies in one company is now the biggest risk. Put IT and the front office on one plan, on one platform, so their agents share data, governance, and results.

79%
Exploring or deploying AI
Front office
Sales, marketing, service, customer success

Base: 608 IT and business decision-makers. Share of organizations exploring or deploying AI agents in each group.

AI adoption surpassed 70% in both IT and the front office

In 2025 an organization was likely to have only one primary function working with AI agents. In 2026, the average organization has nearly four.

In 2025, when we asked which single function was primarily exploring or deploying agents, IT and technical functions were named by 37% of organizations and front office functions (sales, marketing, service, and customer success) by 46%.

In 2026, 74% of organizations have agents in IT and technical functions and 79% in the front office. Both gained more than 30 points in a year, and IT moved up slightly faster.

What it means

IT and the front office are adopting AI agents at nearly the same pace, and in most organizations they are doing it side by side. IT and front office leaders need one aligned AI strategy, and technology that brings their AI initiatives and activities onto a shared platform (see Appendix for more on AI adoption lockstep).

AI agent adoption by function group, 2025 to 2026
IT and technical functions
+37 pts
more organizations exploring or deploying AI agents, 2025 to 2026
37% 2025
74% 2026
IT operations, software development, and R&D and innovation
Front office functions
+33 pts
more organizations exploring or deploying AI agents, 2025 to 2026
46% 2025
79% 2026
Sales, marketing, service, and customer success

Base: 608 (2026, multiple answers; share with agents in at least one function of the group), 564 (2025, single primary function). Point change compares the two years directly, so treat it as directional.

Scaling AI means rethinking how people and agents will work together

AI is changing how work gets done, requiring organizations to rethink how they operate as people and AI agents work together.

Expectations for AI’s role in the workforce are changing. Compared with 2025, more organizations expect AI to create capacity, support growth, and generate new roles. The emphasis is expanding beyond helping employees perform existing tasks toward enabling teams to do more.

Realizing that potential requires clear decisions about how people and AI agents share responsibilities. Organizations need to identify where human judgment is essential, where agents can act with oversight, and where routine tasks can run independently.

This chapter explores how workforce expectations are evolving and how organizations can design effective collaboration between people and AI agents.

The AI Workforce Revolution Isn't About Replacing People. It's About People and AI Working Together

What organizations expect the role of AI agents to be
89% expect AI agents to augment people
9% expect AI agents to reduce headcount
38% create capacity and growth for current staff
34% augment teams without major headcount change
17% create new roles
9% expect AI agents to reduce headcount
2% unsure

Base: 608 IT and business decision-makers.

Workforce expectations are shifting toward greater capacity and new roles

As AI adoption advances, organizations are reconsidering how people and AI agents will work together. In 2025, 48% expected AI agents primarily to augment existing teams, while just 8% anticipated creating new roles as a result of adopting AI.

A year later, expectations are becoming more diverse. The share of organizations expecting AI to create new roles has more than doubled, while 38% expect AI to create additional capacity and support growth. Meanwhile, the share focused primarily on team augmentation has fallen to 34%.

What it means

The shift suggests that organizations are beginning to think beyond using AI simply to help employees do their existing jobs. As agents take on more work, enterprises will need to make deliberate decisions about where that new capacity goes, which responsibilities remain with people, and how roles and teams should evolve around AI. Scaling AI is therefore as much an organizational design challenge as a technology one.

The role expected of AI agents: 2025 to 2026
Create new roles
8% 2025 17% 2026
Create capacity and growth
28% 2025 38% 2026
Augment teams, no major headcount change
48% 2025 34% 2026
Reduce headcount significantly
11% 2025 9% 2026

Base: 608 business and technology decision-makers (2026 Creatio Global Enterprise AI Adoption Survey); 564 business and technology decision-makers (2025 Creatio Global AI and No-Code Adoption Survey).

Expert’s Perspective

AI is here. We need to adapt. What excites us is that AI can take over many of the administrative tasks so salespeople can focus on what they really want to do – being closer to their customers.

Karolina Drózd
Karolina Drózd
Head of Sales Operations, PSI

Agentic AI allows for autonomous IT, which means that organizations can now remove human effort from the lion's share of their workflow. Another way to think about this is that rather than having human work augmented by AI, organizations now have AI work that is moderated by human oversight. The net result is that processes that used to take weeks can now be accomplished in just a few hours.

Rob Garmaise
Rob Garmaise
VP of AI Research, Info-Tech Research Group

Deeper AI adoption is linked to stronger revenue outcomes

Organizations that use AI across a greater share of their workflows are more likely to report increased revenue.

Companies are already seeing measurable business outcomes from AI, including greater productivity, faster decision-making, and improved customer experience. As AI becomes more deeply embedded in the business, its impact extends beyond operational improvements toward revenue growth.

The findings suggest that the value organizations gain from AI evolves as adoption matures. Early deployments can make existing work faster and more efficient, while deeper integration across workflows creates more opportunities for AI to contribute directly to business performance.

This chapter examines the business outcomes organizations are achieving with AI and how deeper adoption is linked to stronger revenue results.

With every AI adoption step, revenue results increase

Share of organizations reporting increased revenue from AI in the past six months
  • Share of workflows automated with AI From under 10% to over half
  • AI governance framework From planning one to fully established

Base: 608 IT and business decision-makers.

Productivity is the first return; revenue growth arrives with maturity

Organizations are already seeing measurable benefits from AI. In the past six months, 61% report increased productivity, 50% faster decision-making, and 43% improved customer experience. More than a quarter (26%) also report increased revenue.

The business impact of AI appears to deepen as adoption matures. Among AI Adoption Leaders, 31% report increased revenue.

What it means

The value of AI may evolve as organizations move beyond initial deployment. Early gains often appear in how work gets done, through greater productivity, faster decisions, and better customer experiences. At more advanced stages of adoption, AI is increasingly associated with revenue growth. The opportunity is not limited to making existing operations more efficient, but to use AI to contribute more directly to business performance.

Where AI is delivering business impact
  • Increased productivity 61%
  • Faster decision-making 50%
  • Improved customer experience 43%
  • Reduced operating costs 30%
  • Improved employee experience 29%
  • Increased revenue 26%
  • Improved compliance 23%

Base: 608 IT and business decision-makers.

Companies scaling AI are 2x more likely to see revenue gains as a positive impact of AI

Revenue results track AI adoption depth for every organization. Among organizations automating under 10% of their workflows with AI, 14% report increased revenue in the past six months. The share rises to 26% at 10%-25% of workflows, 29% at 26%-50%, and 33% above 50%. The agentic share of AI initiatives follows the same curve, from 14%-34%, and so does governance maturity, from 13% of organizations planning a framework to 31% of those running a fully established one.

What it means

The first step past pilots brings the largest single gain: moving from under 10% to 10%-25% of workflows nearly doubles the share reporting revenue, and every further step adds to it. Organizations weighing whether to press on can expect revenue results to follow adoption depth, with governance maturity moving the same curve.

Revenue lift by adoption level, indexed
  • Share of workflows automated with AI
  • Share of AI initiatives that are agentic
  • AI governance framework

Base: 608 organizations, all cohorts, grouped by share of workflows automated with AI (76, 175, 221, 121), share of agentic AI initiatives (111, 167, 187, 115), and AI governance status (38, 114, 267, 176). Index = share reporting increased revenue in each group relative to the first group (1.0x).

What enterprise AI Adoption Leaders do differently

With deeper AI adoption linked to stronger revenue outcomes, the next question is what distinguishes organizations leading that adoption.

AI Adoption Leaders combine extensive AI deployment with more established governance and closer coordination between business and IT teams.

Their approach extends beyond individual deployments. Leaders embed AI across the business, automate a greater share of workflows, and report greater confidence in their ability to scale. They also place emphasis on customer experience and measuring employee outcomes.

This chapter examines the practices that distinguish AI Adoption Leaders from the rest, offering practical priorities for organizations seeking to expand AI adoption and strengthen its contribution to business performance.

Meet the AI Leaders

17% AI Leaders
83% Market Mainstream
17% AI Leaders

AI Leaders account for 17% of surveyed organizations. The remaining 83% make up the market mainstream.

Base: 608 IT and business decision-makers.

How we classified enterprise AI Adoption Leaders

We scored AI adoption maturity on two criteria added together into one AI adoption breadth score from 0 to 16. The two criteria are the number of seven different AI capabilities rated advanced and the number of nine different business functions exploring or deploying AI agents.

AI Adoption Leaders score 10 – 16 points, demonstrating both depth and breadth of adoption. They are advanced across most of the seven AI capabilities assessed and are exploring or deploying AI agents across 6.5 of nine business functions on average, showing that AI is becoming embedded across how their organizations operate.

Seven possible AI capabilities
0-7 points

one for each capability the organization rates as advanced on the four-step scale

  1. Using AI in daily work
  2. Automating core processes with AI
  3. AI agents executing tasks
  4. AI in standard operations
  5. Measuring AI outcomes
  6. Building and configuring AI agents
  7. Overall breadth of AI adoption across the business
Nine possible business functions
0-9 points

one for each business function that is exploring or deploying AI agents today

  1. Sales
  2. Marketing
  3. Service
  4. Customer success
  5. Software development and DevOps
  6. IT operations
  7. Finance and procurement
  8. HR and recruiting
  9. R&D and innovation
16 possible points

Base: 608 IT and business decision-makers.

Three profiles of AI adoption maturity

AI Adoption Leaders make up 17% of all organizations exploring or deploying AI across their organizations. Deployers make up one-third of organizations (33%), and Explorers make up the remaining 50% of organizations.

Leaders

Breadth score 10 to 16

Advanced in most of the 7 AI capabilities, exploring or deploying agents in 6.5 of 9 business functions on average.

  • 63% of workflows run with AI
  • 60% of AI initiatives are agentic
  • 53% already scaling agents across departments
  • 79% very confident about scaling AI

Deployers

Breadth score 6 to 9

Advanced in a few capabilities, with agents in about 4 functions (4.4 on average). AI is in production, not as broadly as the Leaders.

  • 51% of workflows run with AI
  • 44% of AI initiatives are agentic
  • 29% already scaling agents across departments
  • 44% very confident about scaling AI

Explorers

Breadth score 0 to 5

Advanced in 1 or 2 capabilities at most, with agents in one function or 2. Most AI is still in a pilot stage.

  • 21% of workflows run with AI
  • 18% of AI initiatives are agentic
  • 16% already scaling agents across departments
  • 25% very confident about scaling AI

Base: 608 IT and business decision-makers. Breadth score = capabilities rated Advanced (0-7) + functions with AI agents (0-9). Leaders 10-16, Deployers 6-9, Explorers 0-5.

Five things the AI Adoption Leaders do differently

6x

Embed AI across the business

46%

of Leaders describe AI as embedded across the organization, against 8% of others (13% of Deployers and 4% of Explorers). And 53% are already scaling agents across departments, against 29% and 16%.

3x

Deploy AI in more than half of their workflows

48%

of Leaders automate more than half of their workflows with AI, against 14% of others (21% of Deployers and 9% of Explorers). And 41% make more than half of their AI initiatives agentic, against 22% and 10%.

2.5x

Govern AI before they scale adoption

59%

of Leaders run fully established AI governance, against 23% of others (33% of Deployers and 16% of Explorers). And 48% review AI as a formal board agenda item, against 31% and 14%.

1.3x

Adopt for customers and measure people outcomes

38%

of Leaders name customer experience as a reason to pursue agents, against 29% of others (37% of Deployers and 24% of Explorers). And 44% measure employee satisfaction, against 32% and 24%.

2.5x

Scale AI with confidence

79%

of Leaders are very confident they can scale AI, against 32% of others (44% of Deployers and 25% of Explorers). And 42% see the pace of AI as a balanced risk, against 33% and 30%.

Base: Leaders 106, Deployers 198, Explorers 304

Expert’s Perspective

Some of the most useful AI work in any company is already happening, built quietly by individual teams to make their own jobs easier. The job of leadership now is to find those improvements and put them to work across the business, rather than waiting for a top-down strategy. In our own development team, that shift took us from asking what we could do with AI to asking what we can't.

Barney Holmes
Barney Holmes
Senior Vice President, CRM, Bits in Glass

AI transformation isn't a single moonshot. It's a Mars Landing, requiring multiple interconnected missions to succeed. The 17% of organizations leading in AI understand that real transformation isn't about deploying more technology. It's about orchestrating people, governance, and AI as one interconnected system.

Dr. Ja-Naé Duane
Dr. Ja-Naé Duane
Futurist, Best-Selling Author and AI Researcher, Brown University School of Engineering and MIT CISR Academic Fellow

How organizations are approaching AI governance

Established governance is linked to 4x greater confidence in scaling AI.

Organizations must develop trust between people and AI agents, and much of that trust rests on governance.

Nearly three quarters of organizations have a fully or partially established framework, but only 29% call it fully established, and no single governance practice is rated mature by more than half.

Organizations with a framework are more likely to see the risks of moving too fast and too slow as balanced (37%), more than twice the share of those still planning for governance (16%), and fully governed organizations are 4 x more likely to feel very confident about scaling AI.

Governance is often assumed to slow innovation. This section shows the reverse: it comes before the confidence to scale, and the effect is as large inside the Leader group as it is between groups.

Governance is becoming the foundation for scaling AI with confidence

73% have some form of AI governance in place
29% Fully established
44% Partially established
19% Currently developing
6% Planning to create one
2% No plans

Governance is shifting from guardrail to growth enabler.

Base: 608 IT and business decision-makers.

Governance is becoming the foundation for AI at scale

Most organizations have begun establishing AI governance, but relatively few have completed the process. Only 29% report having a fully established governance framework, while 44% have a partially established framework and 19% are still developing one.

Where organizations are on the AI governance journey
  • 29% Fully established
  • 44% Partially established
  • 19% Currently developing
  • 8% Planning or no plans

Expert’s Perspective

AI governance isn't just about guardrails. It's about designing how humans and agents make decisions together. Organizations with established governance are four times as confident about scaling AI, yet only 36% have human review in place. The real challenge isn't how much autonomy we give AI, but how we ensure human judgment remains meaningful.

Dr. Ja-Naé Duane
Dr. Ja-Naé Duane
Futurist, Best-Selling Author and AI Researcher, Brown University School of Engineering and MIT CISR Academic Fellow

Base: 608 IT and business decision-makers.

AI governance is advancing, but maturity remains uneven

Governance maturity is uneven across practices. Organizations are furthest along with more established risk and security controls, like security reviews (46%) and risk assessments (42%). Practices needed to govern AI as its use expands are less mature: only 36% have established human review of AI decisions and 26% have mature AI ethics guidelines.

What it means

As agents take on more responsibilities and AI becomes embedded across more workflows, organizations need governance that covers the full AI lifecycle – from how AI is selected and used to how its decisions are reviewed, monitored, and held accountable.

This means establishing clear usage policies, defining where human oversight is required, and regularly reviewing AI risks and performance. Building these practices early can help governance keep pace with adoption and give organizations greater confidence as they scale AI across the enterprise.

AI governance practices by maturity
  • Security reviews 46%
  • Risk assessments 42%
  • AI usage policies 41%
  • Employee AI training 38%
  • Human review of AI decisions 36%
  • Compliance reviews 33%
  • AI ethics guidelines 26%
  • Vendor governance standards 14%

Base: 608 IT and business decision-makers. Practices question allowed up to three answers.

Governance and confidence scale AI together

The maturity of AI governance is closely associated with how confident organizations feel about scaling AI. Among organizations with fully established AI governance, 71% feel very confident about scaling AI. Among those still developing their governance framework, just 17% say the same. This makes them over 4x more likely to feel very confident about scaling AI.

Mature governance is also more common among AI Adoption Leaders: 36% of organizations with fully established governance qualify as Leaders, compared with just 5% of those still developing their framework.

What it means

Governance is often assumed to slow innovation, but our findings challenge that idea: organizations with more established AI governance are more confident about scaling AI across the enterprise.

At scale, clear policies, controls, and oversight can give organizations the confidence to expand AI into more workflows and grant AI agents more autonomy. The goal, therefore, is not simply to control AI but to create the right conditions to scale it responsibly.

Fully established AI governance by adoption maturity
59%
Leaders
33%
Deployers
16%
Explorers
How confident in scaling AI companies are, by governance establishment phase
  • Fully established 71%
  • Partially established 36%
  • Currently developing 17%
  • Planning to create 11%

Base: 608 IT and business decision-makers. Governance status n = 176 / 267 / 114 / 38; no plans (n = 13) omitted. Leaders 106, Deployers 198, Explorers 304.

Expert’s Perspective

The most important thing to remember about governance is that it needs to be right-sized based upon organizational needs. All too often, we see IT teams that have to peel back previous governance efforts for fear that they have unintentionally become the department of “No.”

Rob Garmaise
Rob Garmaise
VP of AI Research, Info-Tech Research Group

The challenges slowing AI adoption

The challenges that can slow broader AI adoption evolve as organizations move from experimentation to enterprise scale.

Among organizations at the earliest stage of AI adoption, security and compliance, alongside AI skills and organizational readiness are significant.

As adoption matures, skills and organizational readiness become less prominent. At the same time, enterprise-scale requirements move to the forefront: security and compliance rises even further, technology capabilities, commercial models and budgets, and change management become increasingly important.

Scaling AI therefore requires a stronger enterprise foundation than getting started.

AI maturity doesn’t eliminate barriers, it raises the enterprise bar

Barriers that slow broader adoption by maturity
  • Security and compliance 41% → 52%
  • Technology 21% → 37%
  • Change management 25% → 35%
  • Lack of AI skills 40% → 27%

As adoption matures, the challenge shifts from learning how to use AI to building the enterprise foundation required to scale it.

Base: Explorers 304, Deployers 198, Leaders 106.

The barriers to broader AI adoption shift as organizations mature

When asked “As your organization scales AI across the enterprise, which challenges are most likely to slow broader adoption?” organizations at the earliest stage of AI adoption, highlighted security and compliance (41%), alongside AI skills (40%) and organizational readiness (34%).

Among the most advanced adopters, skills and readiness decline to 27% and 21%. At the same time, security and compliance rises to 52%, while technology limitations (37%), commercial models and budgets (35%), change management (35%), and governance (34%) become increasingly important.

What it means

Scaling AI doesn’t eliminate barriers – it changes them. The challenge shifts from building skills and readiness toward establishing the technology, security, governance, and operating model required to run AI at enterprise scale.

Barriers to broader AI adoption by maturity
  • Security and compliance 41% → 52%
  • Technology limitations 21% → 37%
  • Budget 24% → 35%
  • Change management 25% → 35%
  • Lack of AI skills 40% → 27%
  • Organizational readiness 34% → 21%

Base: Explorers 304, Deployers 198, Leaders 106.

Scaling AI across an organization requires an enterprise-grade foundation

As AI expands across the enterprise, organizations need more than strong AI capabilities. They need an enterprise-grade foundation that can support AI across data, security, people, technology, and operations.

When being asked “Which of the following is your biggest barrier to scaling AI across your organization?” the barriers organizations report reflect this broader requirement. Data quality and integration and security are cited by 47% of respondents, while skills and expertise follow at 43%. Technology, change management, organizational structure, governance, and ROI also factor into organizations’ ability to scale.

The pattern is clear: successful AI scale depends on the strength of the broader enterprise environment around it – not on AI capabilities alone.

What enterprises need to address to scale AI
  • Data quality / system integration 47%
  • Security concerns 47%
  • Skills and expertise 43%
  • Technology limitations 28%
  • Change management 27%
  • Organizational structure 24%
  • Governance 20%
  • Unclear ROI 18%

Base: 608 IT and business decision-makers. Multiple responses allowed.

Expert’s Perspective

AI is no longer just a feature layer we bolt onto legacy workflows – it is the new operating system for enterprise. The companies scaling AI successfully aren't just deploying more models; they are completely re-architecting their business from the ground up. AI is shifting from a personal productivity hack into the core engine driving our decisions, operations, and customer value. Ultimately, winning in this era means building an organization that is AI-first by design.

Anatoliy Troshyn
Anatoliy Troshyn
CEO, Banza
What it means

Enterprises should choose AI platforms for where they want to scale, not just where they want to start. Integration, security, governance, and workflow adaptability should be considered from the beginning. An enterprise-ready AI platform can turn these requirements from barriers to scale into foundations for scale.

The AI era is reshaping what buyers expect from enterprise software

In the survey’s software expectations question, 95% selected an option centered on expanding AI, workflows, or workforce capacity without creating new commercial barriers as adoption grows.

This signals a broader shift in enterprise software expectations. AI changes the economics of scale: growth is no longer driven primarily by the number of users or licenses, but increasingly by AI consumption, tokens, agent activity, workflows, and applications.

As adoption expands, organizations need platforms that provide the capacity to scale these dimensions without unpredictable costs or repeated charges that can constrain growth.

95%
expect software to scale freely to support AI
  • 34% Deploy AI without constantly purchasing additional licenses
  • 33% Scale freely across users, AI agents, workflows, and applications
  • 28% Scale as the workforce grows without buying seats
  • 5% No strong preference

Base: 608 IT and business decision-makers.

Expert’s Perspective

Favor outcome-based or flat pricing or consumption caps, so vendor inefficiency doesn't become your cost. If the metrics you measure haven't moved after a set period, treat the AI as a cost rather than an investment, and don't scale it. This doesn't mean not to use AI at all, just that it was applied to the wrong problem – or in a wrong way. Analyze and find a better suited problem to solve using AI.

Thomas Wieberneit
Thomas Wieberneit
CEO and Principal Analyst, aheadCRM
What it means

Enterprise buyers should evaluate software not only by its initial price, but by the economics of scaling AI. The critical question is whether the platform can support more AI consumption, agents, workflows, applications, and business functions while keeping costs predictable and removing commercial friction as adoption grows.

How organizations measure AI value and ROI

AI measurement starts with operational gains, but mature programs broaden the scorecard to business outcomes.

Organizations most often measure AI through productivity gains and time savings. Yet the data shows a measurement gap: the outcomes that motivate adoption are not always the ones organizations track.

As adoption matures, measurement broadens. Leaders are more likely to track revenue impact, employee satisfaction, and innovation outcomes.

This chapter follows that progression: what organizations expect from AI, what they measure, where measurement falls short, and where measurable returns are already appearing.

AI agents are being pursued for more than efficiency — they’re a bet on growth

  • 56% AI agents to ensure increased efficiency
  • 44% AI agents to drive productivity gains
  • 37% AI agents to help build for the future

Base: 608 IT and business decision makers. Share of organizations reporting each result as main reasons why their organization is pursuing an AI agent strategy.

AI programs start with efficiency; mature adopters broaden the value case

Increased efficiency (56%) and productivity gains (44%) are the leading reasons organizations pursue AI agents, followed by building for the future (37%), cost savings (34%), and improved customer experience (31%).

The mix changes with maturity. Productivity is cited by 58% of Leaders versus 39% of Explorers, while customer experience rises from 24% among Explorers to 38% among Leaders. Cost savings, by contrast, falls to 26% among Leaders.

What it means

The value case for AI broadens as adoption matures. Efficiency can justify the initial investment, but enterprise AI programs increasingly need to demonstrate productivity, customer, and longer-term business impact. That makes the choice of success metrics critical from the start.

Main reasons for pursuing AI agents, by profile
AllExplorersDeployersLeaders
Increased efficiency 56% 60% 51% 52%
Productivity gains 44% 39% 44% 58%
Building for the future 37% 35% 40% 40%
Cost savings 34% 35% 37% 26%
Improved customer experience 31% 24% 37% 38%
Operational excellence 30% 30% 27% 34%
Risk reduction 16% 14% 19% 18%
Employee enablement 16% 15% 20% 15%

Base: Explorers 304, Deployers 198, Leaders 106. Multiple answers allowed.

Expert’s Perspective

The question probably needs to be, should we do that with AI rather than can we do that with AI? Because you can probably do most things with AI, but it’s finding the right use case and then really getting that buy-in, realizing the value. I think that builds momentum within an organization and then you let it scale from there.

Christopher Burrows
Christopher Burrows
Director of Operational Systems and Digital Transformation, Veolia

Reasons AI drifts away from what organizations measure

Even when organizations name a clear reason for adopting AI, they do not always measure the corresponding outcome. Among those pursuing productivity gains, 68% measure productivity. Among those pursuing cost savings, 55% measure cost savings.

The gap is larger for strategic goals. Only 56% of organizations pursuing better customer experience measure customer satisfaction; 40% pursuing employee enablement measure employee satisfaction; and 31% building for the future measure innovation outcomes.

What it means

Define the outcome and its metric together. Otherwise, organizations can scale AI activity without being able to demonstrate the value it was intended to create. The biggest measurement opportunity is moving beyond activity and efficiency metrics to a balanced view of business impact.

Who measures AI outcomes that match their reasons
Reason for adoptingMatching metricShare that measure it
Productivity gains Productivity gains
Increased efficiency Productivity gains
Improved customer experience Customer satisfaction
Cost savings Cost savings
Increased efficiency Time savings
Employee enablement Employee satisfaction
Building for the future Innovation outcomes

Base: organizations naming each reason (n shown in order: 268, 339, 187, 207, 339, 100, 226). Reading: of the 268 organizations adopting AI for productivity gains, 68% measure productivity gains.

Expert’s Perspective

As AI matures and scales, measuring AI quality and cost across CX, AI compute, and human effort isn't an audit burden, it's a massive strategic opportunity. It exposes where AI is creating real breakthrough value versus where false resolutions are silently shifting work back onto humans.

A company celebrating a low AI token cost on a customer interaction is blind to the human cost, orders of magnitude greater, generated when a failed fix forces a second call, requiring a human agent to untangle what went wrong, calm an emotional customer, and rebuild lost trust.

Michael Partridge
Michael Partridge
Partner Service Excellence, Deloitte

The next phase of enterprise AI adoption

The report points to a clear shift. Enterprise AI is moving beyond experimentation and becoming part of how organizations operate. As adoption deepens, organizations need to think differently about how work is designed, how AI is governed, and whether their technology can support AI at enterprise scale.

At Creatio, we believe the next phase of AI adoption will be defined by these core principles:

AI should be embedded into workflows, not added as another disconnected tool. People and AI agents need to work across the same end-to-end processes, with the right level of autonomy and human involvement.

AI needs a unified foundation. As AI spreads across functions, organizations need a common environment to build, manage, and use AI agents. This gives business teams room to innovate while IT maintains enterprise-wide visibility and control.

Governance needs to scale with AI. Usage policies, human oversight, security, auditability, and monitoring should be built into how agents operate so organizations can scale AI with confidence.

Enterprise software needs to be built for AI at scale. Organizations need the freedom to expand agents, workflows, applications, integrations, and AI usage without technology or economics becoming a barrier every time adoption grows.

Enterprise AI is entering a new phase. The question is no longer whether organizations will adopt AI, but how they will scale it across the business. That requires a new foundation - one where people and AI agents work together at scale, governance is built in, and technology gives organizations the freedom to expand AI without adding complexity. This is the future of enterprise software we are building at Creatio.

Burley Kawasaki
Burley Kawasaki
SVP Industries, Creatio

About Creatio's Enterprise AI Adoption Survey

Creatio’s State of Enterprise AI Adoption Survey was fielded between August and September 2026 and gathered 608 completed responses from IT and business decision-makers, director level and above, across sales, marketing, customer success, operations, IT, and executive roles in the United States, the United Kingdom, Mexico, and Colombia. Participants received no incentives for participating in the survey.

608

Business and technology decision-makers at director level and above

10+ industries
  • 32% Manufacturing and Distribution
  • 32% Financial Services
  • 14% Technology
  • 8% Professional Services
  • 14% Other
Company size
  • 18% 10,000+ employees
  • 26% 2,500–10,000 employees
  • 40% 250–2499 employees
  • 12% 50 - 249 employees
  • 4% <50 employees

Appendix The Creatio AI Adoption Spectrum

How IT and business teams are aligning on AI adoption

Adoption produces results when IT and the business move together, and only one third of organizations adopt AI in alignment.

The fact that 70% of organizations are beyond experimentation and 79% of boards review AI initiatives does not mean AI adoption is always a highly aligned initiative. In most organizations it is not. This section positions all 608 organizations on the Creatio AI Adoption Spectrum based on whether IT is exploring or deploying agents, how many business functions are doing the same, and whether a governance framework is in place. According to the Spectrum, one third of organizations (and 82% of Leaders) deploy AI that Fits. The rest deploy AI that Sprawls, Stalls, or Sits. We close the report with the Spectrum because adopting AI in lockstep gives organizations the best chance to realize the benefits described in every earlier section.

Introducing the Creatio AI Adoption Spectrum.

The Creatio AI Adoption Spectrum helps organizations align IT and business users so they can deploy AI that Fits the organization while avoiding AI that Sits, Stalls, or Sprawls. While working with early- and late-stage AI adopters, Creatio found that successful adoption depends heavily on deploying AI across IT and business functions with a high degree of lockstep. Below, we place all 608 organizations from the survey on the Spectrum based on the level of AI agent engagement across IT and seven business functions, plus the maturity of AI governance. The Spectrum counts seven business functions because IT operations and software development are combined on the IT axis; the breadth score in Section 02 counts them separately, for nine.

AI that Stalls

Governed, IT engaged, and agents in two or fewer business functions.

AI that Fits

Governed, IT engaged, and agents in three or more business functions.

AI that Sits

No IT engagement and at most one business function exploring agents.

AI that Sprawls

AI in motion without an established governance framework, or business functions running ahead of IT.

One third of organizations deploy AI that Fits

A little over one third (34%) of organizations combine governed AI, IT engagement, and agents in three or more business functions. Another 33% deploy AI that Sprawls, where agents move without established governance, or business functions run ahead of IT. AI Stalls in 22% of organizations where IT runs ahead of business users. And 11% of organizations have deployed AI that Sits.

Leaders are much more likely to adopt AI in lockstep between IT and business users: 82% of them have AI that Fits, and none have AI that Sits. Deployers split (45% Fit, 27% Sprawl), and Explorers Sprawl most (45%).

What it means

Organizations with AI that Sits, Stalls, or Sprawls should create business processes and use AI platforms that align IT and business users on the same governed foundation to improve AI adoption and usher in the business results that follow.

Where each profile lands
  • Leaders with AI that Fits 82%
  • Deployers with AI that Fits 45%
  • Deployers with AI that Sprawls 27%
  • Explorers with AI that Sprawls 45%

Base: 608 IT and business decision-makers. Each dot is one organization, in the colour of its quadrant; positions within a quadrant are illustrative.

The State of Enterprise AI Adoption
The full report, PDF, 53 pages
Download PDF

About Creatio

Creatio is an AI CRM and workflow platform where people and AI agents work together – with no limits on users, agents, or scale. We help midsize and large organizations run customer workflows in the AI era. Headquartered in Boston, MA, with a global team and a large ecosystem of partners, Creatio serves thousands of customers in over 100 countries and automates tens of millions of workflows daily. Genuine care for our clients and partners is at the heart of our DNA.

Bring AI into your everyday workflows

Get in Touch Get a Free Demo
Creatio No-Code Playbook: The Age of AI

Become an AI expert!
Explore more materials

Meet with AI practitioners