North American enterprises are now actively deploying agentic AI systems intended to reason, adapt, and act with complete autonomy.
Data from Digitate’s three-year global programme indicates that, while adoption is universal across the board, regional maturity paths are diverging. North American firms are scaling toward full autonomy, whereas their European counterparts are prioritising governance frameworks and data stewardship to build long-term resilience.
From utility to profitability
The story of enterprise automation has changed. In 2023, the primary objective for most IT leaders was cost reduction and the streamlining of routine tasks. By 2025, the focus has expanded. AI is no longer viewed solely as an operational utility but as a capability enabling profit.
Data supports this change in perspective. The report indicates that North American organisations are seeing a median return on investment (ROI) of $175 million from their implementations. Interestingly, this financial validation is not unique to the fast-moving North American market. European enterprises, despite a more measured and governance-heavy approach, report a comparable median ROI of approximately $170 million.
This consistency suggests that while deployment strategies differ, with Europe focusing on risk management and North America on speed, the financial outcomes are similar. Every organisation surveyed confirmed implementing AI within the last two years, utilising an average of five distinct tools.
While generative AI remains the most widely deployed at 74 percent, there is a notable rise in “agentic” capabilities. Over 40 percent of enterprises have introduced agentic or agent-based AI, advancing beyond static automation toward systems that can manage goal-oriented workflows.
IT operations autonomy becomes the proving ground for agentic AI
While marketing and customer service often dominate public discourse regarding AI, the IT function itself has emerged as the primary laboratory for these deployments. IT environments are inherently data-rich and structured, creating ideal conditions for models to learn, yet they remain dynamic enough to require the adaptive reasoning that agentic AI systems promise.
This explains why 78 percent of respondents have deployed AI within IT operations, the highest rate of any business function. Cloud visibility and cost optimisation lead the adoption curve at 52 percent, followed closely by event management at 48 percent. In these scenarios, the technology is not alerting humans to problems so much as actively interpreting telemetry data to provide a unified view of spending across hybrid environments.
Teams leveraging these tools report improvements in decision accuracy (44%) and efficiency (43%), allowing them to handle higher workloads without a corresponding increase in escalations.
The cost-human conundrum
Despite the optimism surrounding ROI, the report highlights a “cost-human conundrum” that threatens to stall progress. The paradox is straightforward: enterprises deploy AI to reduce reliance on human labour and operational costs, yet those exact factors act as the primary inhibitors to growth.
47 percent of respondents cite the continued need for human intervention as a major drawback. Far from achieving the complete autonomy of “set and forget” solutions, these agentic AI systems require ongoing oversight, tuning, and exception management. Simultaneously, the cost of implementation ranks as the second-highest concern at 42 percent, driven by the expenses associated with model retraining, integration, and cloud infrastructure.
The talent required to manage these costs is in short supply. A lack of technical skills remains the primary obstacle to further adoption for 33 percent of organisations. Demand for professionals capable of developing, monitoring, and governing these complex systems exceeds current supply, creating a self-reinforcing loop where investment increases operational capacity but simultaneously raises human and financial dependencies.
Trust and perception gap
A divergence in perspective exists between executive leadership and operational practitioners. While 94 percent of total respondents express trust in AI, this confidence is not distributed evenly. C-suite leaders are markedly more optimistic, with 61 percent classifying AI as “very trustworthy” and viewing it primarily as a financial lever.
Only 46 percent of non-C-suite practitioners share this high level of trust. Those closer to the daily operation of these models are more acutely aware of reliability issues, transparency deficits, and the necessity for human oversight. This gap suggests that while leadership focuses on long-term overhaul and autonomy, teams on the ground are grappling with pragmatic delivery and governance challenges.
There is also a mixed view on how these agents will function. 61 percent of IT leaders view agentic systems not as replacements, but as collaborators that augment human capability. However, the expectation of automation varies by industry. In retail and transport, 67 percent believe agentic AI will alter the essential tasks of their roles, while in manufacturing, the same percentage views these agents primarily as personal assistants.
Complete agentic AI autonomy is rapidly approaching
The industry anticipates a rapid progression toward reduced human involvement in routine processes. Currently, 45 percent of organisations operate as semi- to fully-autonomous enterprises. Projections indicate this figure will rise to 74 percent by 2030.
This evolution implies a change in the role of IT. As capabilities mature, IT departments are expected to transition from being operational enablers to acting as orchestrators. In this model, the IT function manages the “system of systems,” ensuring that various intelligent agents interact correctly while humans focus on creativity, interpretation, and governance rather than execution.
“Agentic AI is the bridge between human ingenuity and autonomous intelligence that marks the dawn of IT as a profit-driving, strategic capability,” notes Avi Bhagtani, CMO at Digitate. “Enterprises have moved from experimenting with automation to scaling AI for measurable impact.”
The transition to agentic AI requires more than just software procurement; it demands an organisational philosophy that balances automation with human augmentation. Policies alone are insufficient; governance must be integrated directly into system design to ensure transparency and ethical oversight in every decision loop. European organisations are currently leading in this area, prioritising ethical deployment and strong oversight frameworks as a foundation for resilience.
Furthermore, the shortage of technical talent cannot be solved by hiring alone. Organisations must invest in upskilling existing teams, combining operations expertise with data science and compliance literacy.
Finally, reliable autonomy depends on high-quality data. Investments in data integration and observability platforms are necessary to provide agents with the context required to act independently.
The era of experimental AI has passed. The current phase is defined by the pursuit of autonomy, where value is derived not from novelty, but from the ability to scale agentic AI sustainably across the enterprise.
“As organisations balance autonomy with accountability, those that embed trust, transparency, and human engagement into their AI strategy will shape the future of digital business,” Bhagtani concludes.
See also: How the MCP spec update boosts security as infrastructure scales
Want to learn more about AI and big data from industry leaders? Check out AI & Big Data Expo taking place in Amsterdam, California, and London. The comprehensive event is part of TechEx and is co-located with other leading technology events including the Cyber Security Expo. Click here for more information.
AI News is powered by TechForge Media. Explore other upcoming enterprise technology events and webinars here.
