Community
Data and AI leaders explore decision quality, human oversight and shared accountability as organizations prepare for more autonomous AI.
Written by: Camille Prado, Global Editor, CDO Magazine
Updated 2:46 PM EDT, September 11, 2026

Data, analytics, AI and technology executives gather at CDO Magazine’s Columbus Roundtable Exchange at Mastro’s Steakhouse on September 9 for an evening of peer discussion on AI-ready data, trusted context and the agentic enterprise.
That question shaped the conversation at the CDO Magazine Columbus Roundtable Exchange, held September 9 at Mastro’s Steakhouse. Data, analytics, AI and technology executives gathered for an evening centered on “Lead the Agentic Enterprise with AI-Ready Data: Trusted Context for Agents.”
Rather than focusing solely on whether organizations have the data needed to power AI, attendees examined a more consequential question: Can that data provide enough context for an AI agent to make the right decision?
The discussion moved from AI-ready data and decision quality to human oversight, accountability and the growing need for alignment across data, AI, business, legal, compliance and security teams.
Traditional measures of data quality remain important, but agentic AI adds another layer. Data can be accurate and complete while still lacking the context an agent needs to respond appropriately as business conditions change.
For Ibrahim Itani, Vice President, Head of Data, Analytics & AI at Marathon Petroleum Corporation, the conversation reflected the practical questions organizations are now working through as they prepare for more autonomous AI.
“I appreciated the opportunity to participate in the roundtable and exchange perspectives on AI-ready data, trusted context, and what it will take to enable the agentic enterprise responsibly.
“The discussion was engaging, practical, and timely. I enjoyed hearing how others are approaching many of the same challenges around data, AI, governance, and decision quality.”
The shift raises questions that extend beyond data quality itself. Organizations also need to determine how much confidence they have in decisions made by AI agents, what that confidence is based on and what happens when the definition of “good enough” differs across data and AI teams.

Abhishek Singh, Cardinal Health Head, AI CoE, shares insights from his table’s discussion during CDO Magazine’s Columbus Roundtable Exchange.
Greater autonomy also changes the role people play in AI-driven decisions.
Rather than framing the choice as human involvement versus full automation, attendees discussed where agents should be able to act independently, when human judgment remains necessary and who ultimately owns the outcome.
Tina Sebert, Bread Financial Sr. Director, Data Science, left the discussion with a different way of describing that relationship.
“One idea that really stuck with me was shifting the mindset from ‘human in the loop’ to ‘human at the helm’ – where people remain responsible for driving the outcome and AI supports them with better insights and capabilities. But, lots of continued reminders this only works if AI is grounded in trusted, high-quality data. As AI becomes more autonomous, data quality becomes even more critical to the decisions it makes.”
That distinction becomes increasingly important as agents move from providing recommendations to taking actions. Human involvement is no longer simply a checkpoint in a workflow. Organizations need to define where autonomy begins and ends, who can intervene and who remains accountable when an AI-driven decision produces an unintended result.

Sean Erikson, Grange Insurance VP, Enterprise Architecture, IT Strategy, Emerging Tech & Enterprise Integration, leads his table’s report-out during CDO Magazine’s Columbus Roundtable Exchange.
The conversation also surfaced an organizational challenge: the teams responsible for data and those responsible for AI may not always define success the same way.
Data teams have traditionally optimized for qualities such as accuracy, reliability and governance. AI teams increasingly have to consider whether that same data produces an appropriate decision in a particular context.
As new AI use cases emerge, deciding whether data is appropriate for a model or agent can therefore involve far more than a single function. Business leaders, data and AI teams, legal, compliance and security may all have a role to play.
The executives around the tables also discussed the influence of business leaders who own P&Ls and budgets. As those leaders increasingly determine where AI investments are made, data and AI executives have an important role in helping organizations understand what responsible implementation requires.

Ravi Katyal, Cintas Corporation IT Manager – Emerging Technology, reports insights from his table’s discussion at CDO Magazine’s Columbus Roundtable Exchange.
For Columbus leaders, the evening also provided an opportunity to compare what is actually happening inside their organizations rather than relying solely on the broader market conversation around AI.
That exchange extended beyond the structured table discussions, giving executives an opportunity to reconnect with longtime peers, build new relationships and compare experiences across organizations and industries.

Tom Marlow, Liberty Mutual Insurance Global Data & Analytics Executive, and Owen Tak, BMW Group US Analytics COE Manager, connect during CDO Magazine’s Columbus Roundtable Exchange.
For Abhishek Singh, Cardinal Health Head, AI CoE, that openness is one of the strengths of the Columbus community.
“Being part of the Columbus CDO Magazine community is rewarding. The roundtable discussions offer authentic learning that goes well beyond industry hype. It is a trusted environment where we share real stories and I always leave with key insights that I can instantly apply in my own area.”
The peer exchange is particularly relevant as organizations move at different speeds. Some are still defining enterprise AI strategies, while others are working toward systems capable of greater autonomy. Bringing those experiences into the same room allows leaders to compare approaches, pressure-test assumptions and learn from challenges their peers have already encountered.
Technology may have driven much of the evening’s discussion, but Scot Burdette, ABB Global Division CIO, Measurement & Analytics, came away thinking about something more enduring.
“Really enjoyed the CDO Magazine Columbus Roundtable Exchange last night.
“Great to reconnect with long-time friends, catch up with colleagues, and meet new technology leaders from across the Midwest.
“One takeaway: technology will keep evolving, but success still comes down to strong leadership, trusted partnerships, continuous learning, and our ability to adapt.
“Thanks to CDO Magazine and the sponsors for a great evening of conversation, connections, and great food.”
As agentic AI changes what organizations expect from their data, the questions facing data and AI leaders are expanding with it. Data readiness is increasingly about more than whether information is accurate and available. It is also about whether organizations have the context, governance, human judgment and cross-functional alignment needed to trust what AI does with it.
Special thanks to Cyera, Immuta, BMC Software, and Fortinet for making this event a success.
Executives attending the CDO Magazine Columbus Roundtable Exchange included:
*Scot Burdette, Ibrahim Itani, Tom Marlow, Mark Marshalek, Sanjay Saxena and Payal Thakur are members of the CDO Magazine Global Board.