IBM Study Reveals Challenges in Scaling AI Across Enterprises
On June 8, 2026, the IBM Institute for Business Value published a comprehensive study highlighting the growing challenges that Chief Information Officers (CIOs) and Chief Technology Officers (CTOs) face as artificial intelligence (AI) transitions from experimental phases to widespread deployment. The study indicates that two-thirds of surveyed technology leaders are held accountable for AI systems they do not fully control, raising concerns about governance and operational risks as organizations prepare for a significant increase in AI agent deployment.
Accountability Without Control
The global survey, which included responses from 2,000 C-level technology executives across various industries and geographies, revealed alarming trends regarding accountability in AI governance. A staggering 70% of respondents noted that technology teams are deploying new technologies at a pace that IT departments struggle to monitor effectively. This lack of visibility is particularly concerning given the anticipated 38% increase in AI agents by 2027.
Despite the urgency to scale AI initiatives—driven largely by CEO mandates—only 11% of respondents feel fully prepared for the forthcoming surge in AI deployments. Governance structures are failing to keep pace, with 77% of organizations reporting that their current capabilities cannot manage the rapid adoption of AI technologies.
Matt Lyteson, CIO at IBM, emphasized the need for organizations to rethink their approach to AI governance: “The challenge now is scaling AI systems that operate continuously and autonomously within governance models designed for a far slower environment. It’s about embedding control and visibility from the start.”
Operational Risks and Security Concerns
The study also identified escalating operational and security risks associated with increased AI adoption. Organizations relying on manual governance mechanisms reported a higher incidence of problems as they scaled their AI systems. Specifically, those that integrated control directly into their AI frameworks experienced 25% fewer incidents compared to those using manual methods.
Concerns about security and compliance emerged as significant barriers to scaling AI effectively, with 59% of tech CxOs citing these issues as top challenges. Last year alone, surveyed organizations faced an average of 54 incidents involving AI agents requiring human intervention due to unintended or harmful outcomes. Alarmingly, 17% of these incidents were classified as high severity, necessitating extensive containment efforts.
- 37% resulted in data exposure or security breaches.
- 33% caused cascading system failures.
- 17% triggered compliance issues.
Redesigning Control for Better Outcomes
The study highlights a clear correlation between effective governance structures and improved performance outcomes in organizations deploying AI. Currently, spending on AI is projected to rise dramatically—from just under 15% of IT budgets in 2025 to nearly 25% by 2027—a staggering increase that raises stakes for CIOs and CTOs alike.
Despite this impending growth in investment, many organizations remain ill-prepared; 84% have not fully operationalized financial management related to AI initiatives, while 85% lack real-time visibility into their spending on these technologies. The findings suggest that organizations embedding control into their AI systems see substantial benefits:
- They deploy up to 16 times more AI agents than those relying on manual governance.
- Their operating margins are approximately 18% higher.
- They spend four times less of their overall budget on AI initiatives.
Moreover, companies demonstrating strong financial discipline can deploy 2.4 times more AI agents without increasing their IT budgets and are three times more likely to feel fully prepared for scaling their operations effectively. Organizations that prioritize adaptability—ensuring workloads remain portable and models can be easily replaced—reported a return on investment (ROI) from AI that was 10% higher than their less adaptable counterparts in 2025.
The Path Forward
The full findings from the IBM study provide critical insights into how technology leaders can navigate the complexities associated with scaling AI across enterprises. Recommendations include redesigning organizational structures to enhance speed, control, and investment strategies tailored for rapid technological advancement.
This research underscores the necessity for businesses not only to accelerate their adoption of artificial intelligence but also to implement robust governance frameworks capable of managing associated risks effectively. As organizations prepare for an era where reliance on automated systems will only grow stronger, establishing clear accountability and control mechanisms will be paramount.
What This Means
The implications of this study extend beyond mere statistics; they highlight a crucial turning point for enterprises looking to leverage artificial intelligence effectively. As technology leaders grapple with accountability without control, it becomes clear that success hinges on proactive governance strategies designed for scalability. Organizations must invest not only in technology but also in building frameworks that ensure responsible use while maximizing operational efficiency. The future landscape will favor those who can balance innovation with oversight—an essential consideration as businesses navigate this transformative era.
For more information, read the original report here.
