CIO 100 Leadership Live Atlanta Coverage: AI Spending Enters a Reckoning Phase as Enterprises Demand Returns on Early Bets

After two years of heavy investment in artificial intelligence, corporate technology budgets are hitting a moment of truth, with executives facing mounting pressure to show that early AI commitments are generating measurable returns rather than accumulating as sunk costs.

That market reality framed nearly every conversation at the CIO 100 Leadership Live conference held March 5 at the Westin Buckhead Atlanta. The event drew chief information officers and senior technology leaders from industries spanning transportation, manufacturing and financial services to healthcare, retail, professional sports and higher education. Across sessions, a consistent picture emerged of an enterprise AI market shifting from a growth-at-all-costs posture to one defined by governance discipline, data accountability, and business justification.

"The technology continues to evolve very quickly," said Jeff Baker, U.S. tech managed services leader at PwC, who addressed attendees and co-hosted a separate executive roundtable during the event. "But organizations still have to connect these capabilities to measurable outcomes and ensure that they are implemented in a way that is sustainable."

Leadership Must Evolve Alongside the Technology

The morning opened with a keynote from Keith Fulton, chief data officer at Jack Henry and author of "Maxing Out: How to Get the Most out of Yourself and Your Team," published in October 2024. The book, which is gaining traction in CIO and IT leadership circles, draws on Fulton's own trajectory from software engineer to senior executive to argue that technical expertise alone is insufficient preparation for enterprise leadership.

Fulton told attendees that the skills required to drive AI transformation at scale are fundamentally different from those that earned most technology leaders their first promotions. Where early career success rewards technical depth, leading large organizations through AI adoption demands capabilities in budgeting, organizational design, cross-functional collaboration, and communicating complex change to executive peers and boards.

His central argument was for technology leaders to shift from a mindset of control to one of enablement. In large organizations navigating rapid change, he said, the leaders who succeed are those who build cultures where people understand the real impact of their work, trust their teams to execute, and maintain the personal discipline required to sustain performance over time. Fulton also stressed that long-term leadership effectiveness requires work-life balance, arguing that leaders cannot be dependable to their organizations if they are not dependable to themselves.

Avoiding the POC Pit

A panel titled "Real Leaders, Real Challenges" followed, featuring Marianne Johnson, chief product officer at Cox Automotive; Justin Mennen, chief information and technology officer at Shake Shack; and Afshean Talasaz, former chief technology and data officer at Colonial Pipeline. Moderated by Dan Roberts, host of the Tech Whisperers podcast, the session drew on firsthand accounts of technology transformation under real business pressure.

Panelists described moving beyond isolated AI experiments toward integrating the technology into core operations, a shift that demands organizational will and acumen as well as technical capability.

The afternoon session on the business case for AI sharpened that focus. Chris Crist, CIO and senior deputy general manager at Hartsfield-Jackson Atlanta International Airport, and Bryan Muehlberger, global CIO at Lumiyo, discussed how their organizations are building frameworks to measure AI's return on investment and avoid the trap of perpetual piloting, where projects never advance beyond the proof-of-concept stage into full operational deployment.

The airport, one of the busiest in the world, is building an innovation team with an eye toward using AI to improve both operational efficiency and passenger experience. The challenge, participants noted, is moving fast enough to stay relevant while not outpacing the organization's ability to govern what it deploys.

Knowledge Management Becomes a Strategic Asset

An early morning roundtable hosted by Unisys tackled a challenge that would echo throughout the rest of the day, namely that organizations often underestimate how much their AI outcomes depend on the quality and structure of their enterprise knowledge.

Joel Raper, chief commercial officer at Unisys, said many companies are discovering that knowledge management systems built for a previous era of IT are ill-equipped to support AI-driven automation and decision making.

"Knowledge management becomes the enabler for so much more of the business," Raper said. "You have to put the guidebook in place. If you want AI to act on something, it has to understand how the organization actually works."

Absence of Data Governance Emerges as the Hidden Bottleneck

A midafternoon panel on data strategy reinforced those themes, surfacing evidence that AI ambitions routinely outrun the quality of the data supporting them across industries.

Anjali Arora, chief technology officer at Perforce Software; Ben Pivar, senior vice president at Fanatics; and Jikin Shah, senior vice president of technology infrastructure at RBC, joined Roberts for a discussion on building enterprise data readiness.

Panelists described organizations discovering, often mid-deployment, that their data foundations were not ready for the demands AI places on them. Fragmented data, inconsistent governance, and unclear ownership are slowing AI initiatives that looked promising on paper.

The message to peers in the room was that investing in data stewardship is not just a prerequisite for AI adoption that can be deferred. It is the work itself.

Venture Capital Offers a Long View on AI Anxiety

In a panel moderated by TechCrunch reporter Dominic Madori Davis, three venture capital investors shared what enterprise technology leaders can learn from how the investment community evaluates AI risk.

Lisa Calhoun, founding managing partner at Valor Ventures; Shila Nieves Burney, founder and managing partner at Zane Venture Fund; and Amy Lambert, CFO at SteelSky Ventures, pushed back on what they described as misplaced fear about an AI investment bubble.

They drew a distinction between inflated startup valuations and the durability of the underlying technology. AI integration into enterprise operations, they argued, is not a speculative bet but an inevitability, and the more productive question for organizations is how to manage that integration effectively.

On the question of evaluating AI vendors, the panel steered away from valuation metrics as a primary measure of reliability. The more meaningful indicator, panelists said, is whether the investors behind a given company have the appetite and resources to support it through market turbulence. 

The panel also addressed how enterprises should assess early-stage AI startups. Panelists argued that product maturity at the time of evaluation is less important than the founding team's clarity of vision and capacity to navigate adversity. While products evolve, the conviction and direction of leadership are harder to change and therefore more predictive of outcomes.

Southeast as Emerging Tech Ecosystem

A quieter theme running through the day was the role of Atlanta and the broader Southeast in the national technology landscape. Several speakers noted that while the region has historically lagged the West Coast in venture capital activity and corporate-VC collaboration, that gap is narrowing.

Calhoun pointed to a growing number of AI startups emerging from the Southeast and encouraged the CIOs in the room to engage more directly with the local venture community, a relationship she said remains underdeveloped compared to counterparts on the coasts.

A New Phase, Not a New Cycle

Across sessions, speakers resisted framing AI as simply another technology wave to manage. Baker, speaking at a PwC-hosted roundtable later in the evening, argued that organizations which succeed will treat AI not as an IT project but as a multidisciplinary leadership pillar that connects strategy, operations, and accountability at every level of the enterprise.

That framing aligned closely with the message Fulton had delivered in the morning. The skills that made someone a strong technologist, he had told the room, are not the same skills that will make them an effective leader through the changes ahead. Developing those capabilities, he argued, is not optional.

###