Sapient Slingshot wins Technology Breakthrough of the Year.
Sapient Bodhi ranked #1 for deep research.
Meet Sapient Sustain: context-aware AI for complex IT operations.
3M lines of COBOL → clear specs in 8 weeks.
use AI regularly
expect to scale AI in the next year
are equipped to deliver
Inside most large enterprises, two things are now true at once: AI has changed how work actually gets done, and almost nothing about how the enterprise operates has changed to match. Teams ship code faster, draft reports in minutes and run customer workflows that used to take weeks, while budgeting cycles, governance reviews and the systems underneath them still move at the pace they did 10 years ago.
This is AI theater, where leaders can point to adoption but not outcomes. To get the most value out of AI, enterprises need to dismantle the structural bottlenecks that stop it from having a real, business-wide impact. The organizations making progress aren't necessarily the ones adopting more AI. They're the ones learning to operate their workflows and infrastructure differently because of it.
"
In order to actually scale AI, you have to think about how AI is going to fundamentally redesign how people and platforms in an organization interact to create value, not just very simply apply AI to either an existing workflow or an existing area of the business.
"— Nigel Vaz
CEO, Publicis Sapient
73 percent of enterprises use AI regularly or in most processes, but only 10 percent say it is core to how their business operates
38 percent say AI is fundamentally changing how their business operates, and 18 percent say AI is fully integrated across the enterprise
Leaders are twice as likely to blame the way their organization runs (22 percent) as the capability of AI itself (11 percent)
65 percent of leaders expect significant progress in scaling AI over the next 12 to 24 months, but only 15 percent say their organization is fully equipped to support that
The most-cited limits on scaling AI are integration across systems (38 percent), data fragmentation (36 percent) and governance and control (29 percent)
The enterprises pulling ahead are making three shifts: modernizing the systems that slow the business, coordinating AI across the enterprise and rethinking operational resilience
AI is showing up everywhere except the operating model. Seventy-three percent of enterprises use AI regularly or in most processes, yet only 10 percent say it is core to how the business operates (Figure 1).
AI adoption inside large enterprises no longer looks experimental. Across industries, teams already use AI inside everyday work. They generate code, automate reporting, personalize customer experiences and speed up delivery cycles that used to take months.
In many organizations, AI changed the pace of execution before leadership fully realized what was happening. Employees now expect work to move faster because, in many parts of the business, it already does.
Widespread use doesn’t necessarily translate into enterprise-wide impact. Most leaders (62 percent) do not yet say AI is fundamentally changing how the business operates, and the share varies significantly by region. In the U.K., 51 percent say it is, compared with 24 percent in France (Figure 2).
Figure 1: AI is part of everyday work
The majority of employees are using AI regularly, even where it isn’t yet core to how they operate.
Figure 2: Business-wide transformation is uneven
Some markets are redesigning the business around AI. Others are still layering AI onto legacy operations.
Leaders believe the technology works, but they don't believe their organizations do
One of the clearest shifts in the survey appears when leaders assess what limits their ability to realize AI value today. They don’t have an issue with the technology itself: nearly half (47 percent) say AI is already fully capable of fulfilling current business needs, and only 11 percent say it is not capable yet. Instead, the bottleneck is structural. Forty-two percent say their organizations are not structured to capture the value AI can already create (Figure 3).
That distinction reflects organizational complexity rather than technological maturity. In highly regulated, operationally dense environments, legacy systems and siloed structures drag down teams and become harder to work around, turning what should be a seamless deployment into a coordination crisis.
Figure 3: AI's capabilities aren't the issue
Enterprise leaders believe in their organization's AI capabilities.
AI changed the pace of business. Enterprise systems did not.
The data is clear: teams are already using AI in their day-to-day work. But changing teams' tools is very different from changing how enterprises operate, especially when most organizations still rely on systems, models and structures built before AI. Many companies adopted the technology much faster than they could coordinate around it.
The enterprise itself has become the bottleneck. Leaders cite integration across systems (38 percent) and data fragmentation (36 percent) more than any other limit, and they are twice as likely to blame how their organization runs as the capability of AI.
Across regions, respondents increasingly describe friction tied to coordination, governance, workflow integration and organizational readiness rather than AI capability itself.
This bottleneck shows up in the day-to-day experience of trying to scale AI across a large organization:
AI efforts spread across functions without a shared strategy to connect them
New projects launch faster than organizations can build institutional knowledge from them
Leadership struggles to maintain visibility across a growing mix of tools, workflows and use cases
Teams spend more time managing complexity than leaning into what already works
These hurdles may be invisible during early adoption, but they become much harder to ignore once organizations try to scale.
AI workflows scaled faster than enterprises can integrate them
Even though AI usage is now widespread, most enterprises haven't been able to operationalize AI consistently across the organization. Fewer than one in five respondents (18 percent) say AI is fully integrated across the enterprise (Figure 4). The funnel narrows at every step: 83 percent say AI is used regularly, in most processes or as core to operations; 68 percent say it is used regularly or is core to how the business runs; 38 percent say it is fundamentally changing how the business operates; and 18 percent say it is fully integrated across the enterprise.
Figure 4: What AI usage looks like in the business today
Organizations are starting to embed AI into workflows, but most enterprises still lack the systems to scale it coherently.
Fragmentation became the default operating model
The problems slowing AI scaling are operational: systems don't connect and data remains fragmented (Figure 5). Without integrated systems or shared data, AI will always be limited in scope and unable to deliver enterprise-wide value. Respondents most often cited integration across systems (38 percent), data fragmentation (36 percent), governance and control (29 percent), organizational silos (26 percent), lack of workflow integration (21 percent) and inconsistent tools (19 percent).
Figure 5: What limits AI scaling most
Respondents point to a number of issues that block their ability to scale AI across the enterprise.
"
So when you think about AI more broadly at an enterprise level, you have to think about the systems and the operating models that it is going to interact with. And that interaction is typically where these points of failure occur.
"— Nigel Vaz
CEO, Publicis Sapient
Executive ambition keeps rising while operational readiness lags behind
Globally, respondents have high hopes for AI. Sixty-five percent expect significant progress in scaling AI across the business over the next 12 to 24 months (Figure 6). Yet only 15 percent say their organizations are fully equipped to support those expectations today, a 50-point gap. And just 30 percent say executive expectations are fully aligned with operational reality . That combination suggests many enterprises are accelerating AI expectations faster than organizations can adapt.
Figure 6: Scaling versus readiness
Enterprises expect massive AI scaling without operational readiness.
The organization itself has become the constraint
One of the clearest findings in the survey is how often respondents point back toward the enterprise operating model itself as a hurdle. More than one in five respondents (22 percent) say "the way our organization runs" is the primary blocker to AI success, twice the share who name AI capability itself (11 percent; Figure 7).
That finding changes the conversation. Most enterprises already have access to the models, and many already run pilots across the business. But the harder problem is managing how the business actually runs, both in terms of building the right infrastructure and equipping teams with the right workflows. Organizations now have to coordinate around a technology that speeds up processes faster than they were built to handle.
Figure 7: Technology isn't the issue anymore
Enterprise leaders are twice as likely to blame how their organization operates as they are to blame AI itself.
Enterprises can't keep operating like they always have
As AI spreads across teams and functions, enterprises have started running into the limits of their systems and structures. What began as a technology rollout has quickly snowballed into an organizational quagmire.
Enterprises can no longer organize and behave the way they did before AI arrived. Instead, if they want to translate their AI investments into enterprise-wide impact, they're going to need to make broader operational changes around AI.
"
You have to be really intentional about redesigning your operating model around AI rather than simply deploying it. In the context of how things work today, you need to be thinking about this whole notion of how a digital workforce and a human workforce can start to work together.
"— Nigel Vaz
CEO, Publicis Sapient
The next enterprise divide will be about adaptation. Most large enterprises have access to the same AI models, cloud providers and copilots. The bigger difference increasingly comes down to how organizations adapt around them.
Some enterprises are redesigning workflows, modernizing infrastructure and coordinating AI across the business in more systematic ways. Others are layering AI onto fragmented operating environments that still rely on slower governance models, disconnected systems and legacy delivery structures. That gap is starting to shape the next phase of enterprise competition.
Many enterprises still mistake AI adoption for transformation
The survey shows a growing gap between AI usage at the team level and actual enterprise transformation. Though teams may be using AI in various functions, that usage isn't having a significant, enterprise-wide impact yet. Forty-nine percent say AI is just starting to change parts of the business, 13 percent say AI is widely used but has not changed how the business operates and 38 percent say AI is fundamentally changing how the business operates (Figure 8).
Figure 8: AI adoption is outpacing enterprise transformation
AI is already embedded into how work gets done, but most enterprises still haven't redesigned the business around it.
That gap increasingly reflects operational fragility underneath the enterprise itself. Respondents consistently point to integration, governance and silos as major barriers (Figure 5). These are all organizational issues, and they all stem from the enterprise's inability to adapt. Disconnected systems and slow workflows have become increasingly difficult to scale around, and that pressure has made it clear that enterprises need to make operational shifts.
The enterprises pulling ahead are adapting their operating model to AI
Some organizations are beginning to move beyond AI deployment and into operational adaptation. That distinction matters because individual functions are no longer the main enterprise bottlenecks. The bottleneck is the lack of coordination between them. Product, risk and compliance may all use AI tools independently, but the business still slows down as decisions stall every time work crosses a functional boundary. This is something AI can't fix on its own.
The enterprises that are able to adapt in the AI era are changing how their business operates. Instead of deploying AI team by team, they are rebuilding how work moves through the business itself. They are increasingly focusing on three shifts.
Modernization
What adapted enterprises do differently: They modernize systems earlier and make critical knowledge accessible across teams and AI workflows, and they embed operational context, rules and institutional knowledge directly into AI systems.
See it in practice: A global bank used Sapient Slingshot to analyze nearly three million lines of COBOL and turn them into verified specifications in eight weeks, cutting analysis per feed from 35 days to five.
Coordination
What adapted enterprises do differently: They connect workflows across functions so decisions move through the business with fewer handoffs and less rework, and they create enterprise-wide strategies that connect tools, workflows and priorities across the business.
See it in practice: A global consumer goods leader used Sapient Bodhi to connect content work across dozens of brands and markets. It created more than 700 assets in two months, reused 60 percent of them across brands and cut production cycles from weeks to days.
Resilience
What adapted enterprises do differently: They create visibility across AI activity, workflows and outcomes so teams can coordinate around a shared view, and they reduce operational debt, simplify processes and remove repetitive work that slows execution.
See it in practice: Nissan used Sapient Sustain to move IT operations from reactive support to a proactive model without replacing its existing stack. Operational costs fell 40 percent, and 62 percent of issues are now resolved the same day.
These shifts may sound operational on the surface, but they determine whether AI creates business-wide outcomes or simply adds more chaos to already fragmented enterprises.
"
I think when you start to think about any major technology shift, you have to start to get the foundation right. You have to recognize that the enterprise today was not built for a fully autonomous agentic architecture. That journey essentially means that you can start to lay out a very clear path that goes from 'how do we modernize legacy,' to 'how do we move into the cloud,' to 'how do we actually establish a new architectural stack across the technology layer,' and then start to move to the human aspects.
"— Nigel Vaz
CEO, Publicis Sapient
Organizations must adapt to succeed
Many organizations have already embedded AI into workflows and daily operations. The next challenge involves redesigning the enterprise itself around AI.
Some enterprises are adapting faster than others. More and more, the difference comes down to modernization, operational coordination and organizational adaptability rather than technology access alone. The enterprises that reorganize around AI-era speed fastest may shape the next era of enterprise competition.
"
I think you have to get really focused on value pools in order to think about how you can reimagine them in the context of real goals, like improving margins or speed, or resilience, or customer outcomes. Because, ultimately, the scale and strategic choices that you make follow economics, not novelty or technology.
"— Nigel Vaz
CEO, Publicis Sapient
The dominant enterprise AI story assumes organizations naturally adapt once AI enters the business. But organizations are finding the opposite is happening. AI is already changing how work gets done, but it is not fundamentally shifting organizations to create value. These shifts simply can't happen when the systems surrounding AI operate the same way they did before.
That gap increasingly defines enterprise AI execution. Many organizations now run 21st-century technology inside 20th-century operating models. Teams work at the speed of AI while tripping over governance systems, budgeting cycles, delivery structures and legacy infrastructure built for another era.
As AI becomes more deeply embedded into operations, enterprises face growing pressure to modernize their infrastructure, coordinate workflows, integrate systems and operate with far greater organizational flexibility than most legacy environments were designed to support. The enterprises adapting fastest are focused on three operational shifts:
Those shifts mark the difference between real operational change and AI theater. AI needs to be more than simply visible across the business. It needs to be integrated into the enterprise's very core.
AI has a way of revealing how slow, fragmented and operationally inflexible enterprises already were before it arrived. But it also gives enterprises permission to reorganize systems, workflows and ways of operating that have needed to change for years. For many organizations, this may be the first real opportunity in decades to stop working around enterprise complexity and start removing it.
Publicis Sapient's platforms are built to run AI at enterprise scale and integrate it into the business. They equip enterprises to make the three operational shifts they need to see a return on their AI investments.
Modernization. Slingshot helps enterprises modernize software delivery and legacy infrastructure so teams can build, ship and scale faster without dragging decades of technical debt behind them.
Coordination. Bodhi helps organizations coordinate AI agents, workflows and enterprise systems with the help of operational and industry context that is embedded directly into the platform.
Resilience. Sustain helps businesses stay resilient as AI accelerates operational complexity, using agent-driven operations to automate incident response and reduce overhead.
"
Publicis Sapient, as an enterprise technology company, is built around a series of platforms: Slingshot, Bodhi and Sustain. Each of these individually solves part of the problem. But together, they provide an incredible AI platform capability and organization where they can really enable the shift of significant workflows around the business to accelerate how intelligence is brought to the organization.
"— Nigel Vaz
CEO, Publicis Sapient
Publicis Sapient surveyed 1,550 enterprise AI decision-makers across the United States, United Kingdom, France, Germany, Australia and the United Arab Emirates from April 29 to May 14, 2026. Respondents work at organizations with at least 500 employees and $100 million in annual revenue, and are responsible for evaluating, influencing or selecting enterprise AI technologies and platforms. The sample included 575 respondents in the United States, 300 in the United Kingdom, 250 in France, 215 in Germany, 150 in Australia and 60 in the United Arab Emirates (small base). Fieldwork was conducted by Protégé on behalf of Publicis Sapient.
