At the Gartner IT Symposium/Xpo 2024, the spotlight was on one of the most pressing challenges faced by CIOs and technology leaders today, artificial intelligence. The opening keynote, delivered by Mary Mesaglio and Hung LeHong, highlighted two concurrent AI races that every organization must contend with: the tech vendor race, driven by relentless innovation, and the internal race to deliver safe, scalable AI outcomes.

The presentation emphasized that while the pace of innovation outside organizations is breakneck, the real race for CIOs lies within, navigating AI adoption to ensure real business value without falling prey to hype.

In 2025, CIOs will continue to face the critical task of balancing these dual races, determining their own pace of AI adoption, and delivering sustainable outcomes that match their business goals.

The Dual AI Race: Innovation vs. Outcomes

Mesaglio and LeHong, made it clear that CIOs are not expected to keep up with the industry’s frantic AI development pace. Instead, they need to focus on delivering meaningful AI outcomes. The tech vendor race, where innovation is happening at an unprecedented speed, can often overwhelm IT leaders with the pressure to adopt every new AI tool.

However, CIOs are urged to focus on their own internal race—the race to deploy AI at a pace that fits their organizational needs.

“Because of the relentless innovation happening in the tech vendor race, CIOs feel like they are always living the hype, while the reality of their AI outcomes race—how tough it is to get value—makes it feel like they are also in the trough,” said Mary Mesaglio, Distinguished VP Analyst

The real challenge lies in balancing innovation with practical, scalable AI outcomes that deliver real business value.

AI Pace: Steady vs. Accelerated

The speakers introduced two distinct AI adoption paces: AI Steady and AI Accelerated, emphasizing that CIOs must determine which pace is appropriate based on their industry, business goals, and AI ambitions.

• AI Steady Pace: For organizations in industries where AI isn’t transforming the core business yet, or for those with modest AI ambitions, a more measured approach is suitable. Here, the focus is primarily on improving employee productivity. Many enterprises aim to harness AI for simple gains, such as boosting employee efficiency. However, even this can be challenging, as integrating AI tools into daily workflows remains a hurdle.

• AI Accelerated Pace: In contrast, for industries being rapidly transformed by AI—such as manufacturing, healthcare, and financial services—an accelerated AI strategy is essential. These organizations aim to drive not just productivity but also process improvement, customer experience enhancements, and business model innovation. To achieve these goals, CIOs must push beyond employee productivity and manage a portfolio of AI benefits at scale.

CIOs must ask themselves, what are the key benefits they seek from AI? Those aiming primarily at productivity gains may opt for a steady pace, while those seeking deeper benefits like new revenue streams or innovation will need to accelerate.

Business Outcomes: Matching Expectations with Reality

The path to AI productivity is neither quick nor easy. While employees may be eager to experiment with AI tools, getting them to integrate these tools into their daily work remains a challenge.

This concept of productivity leakage where time saved by AI doesn’t necessarily translate into business value was central to the discussion.

In some cases, junior employees benefit more from AI tools, as they allow them to close the gap in experience faster. In contrast, highly experienced workers may find the same AI tools less impactful. This underscores the importance of matching the complexity of the task with the experience level of the user when aiming for productivity gains. AI benefits are not distributed evenly across all employees.

For organizations running at an accelerated pace, AI is not just about employee productivity but unlocking wider business benefits.

Managing AI Costs: A Hidden Risk

While AI offers significant benefits, Mesaglio and LeHong warned CIOs that AI investments come with steep and unpredictable costs. From inference costs to data preparation, the expenses associated with AI can easily balloon. The keynote stressed the need for cost transparency and real-time monitoring, particularly for organizations pursuing AI at an accelerated pace. LeHong pointed out that some organizations avoid millions in AI costs by opting for API integrations over packaged AI products, a strategy that may prove invaluable for cost-conscious IT leaders.

To mitigate cost risks, CIOs must move beyond traditional proof of concept models and adopt a proof of value mindset, ensuring that AI projects not only work technically but also deliver measurable financial benefits and cost scalability.

Tech Sandwich: A New Approach to AI Integration

One of the standout concepts from the keynote, and what looks to become the common thread throughout the conference, was the idea of a “tech sandwich” replacing the traditional tech stack in an AI-driven world.

With AI being embedded in an increasing number of enterprise applications, and data coming from a variety of structured and unstructured sources, IT leaders must rethink their architecture. The tech sandwich layers AI and data in a decentralized manner, reflecting the reality that AI is now everywhere in the enterprise from CRM systems to HR tools to in-house developments.

CIOs are urged to design their tech sandwiches carefully, balancing centralized IT control with the decentralized nature of today’s AI applications.

In this model, TRISM technologies (Trust, Risk, and Security Management) become critical, acting as the middle layer that ensures AI governance, risk monitoring, and security are maintained as AI tools proliferate across the organization.

This approach enables CIOs to manage AI complexity while maintaining flexibility and control.

The Wrap

As CIOs look ahead to 2025, the future for AI is both exciting and somewhat daunting.

The duality of innovation and outcomes will continue to push IT leaders to make strategic decisions about how and when to deploy AI. The choice between an AI steady or AI accelerated pace depends on the specific needs of the organization, but the key is to remain focused on delivering real business value.

Whether it’s managing AI costs, integrating AI into existing systems, or navigating the behavioral implications of AI on employees, CIOs must take a measured approach to ensure AI success.

In this race, it’s not about keeping up with the latest technology, it’s about setting your own pace and winning the race.