As organizations continue to make significant investments in AI, many CIOs are still working to understand how those investments will have a measurable impact on the business. At the heart of that challenge is a shift in the way we approach AI, from isolated experiments to enterprise-wide execution. In this conversation, Jeff Baker, Technology Managed Services Lead at PwC, shares how organizations can move beyond early-stage use cases and start realizing meaningful outcomes.
Jeff Baker, PwC Technology Managed Services Lead
CIO.com: Many CIOs are investing in AI, but are not necessarily seeing a return on their investment yet. What does it take to move from investment to actual innovation?
Jeff Baker: A few things. I don’t think many of our clients are thinking deeply enough about the impact of AI and the possibilities there are. One of the things we encourage them to do is to get out of the experimentation stage, out of the back office, out of the cottage industry, and actually partner directly with businesses to find more impactful ways to use technology that drive business outcomes, rather than just being a showcase of cool technology.
There are a lot of skunkworks projects that look fun but aren’t necessarily profitable from an impact standpoint. The more teams you have with AI engineers and internal technology enthusiasts, the more meaningful the results will be.
CIO.com: You’ve said that AI requires structural change, not just experimentation. What are the most important operational shifts that CIOs should make?
Jeff Baker: I think of AI in two basic categories. There’s something I call “citizen-driven AI.” We’re giving a lot of really cool tools to people in companies, and they’re doing interesting things with it. They create chat programs to organize their inboxes and respond to RFPs and other “everyday” tasks.
However, there are also more durable agent-based models that have a greater business impact but require more investment. Strong teamwork between IT and the business is critical here to define what the outcome should be.
There’s also a lot of refinement that comes with it. Is it durable? Are you safe? Are you thinking about bias? How do you curate it? Who owns the ongoing management and observability of these agents once they are deployed?
Security and data management will be key. An agent’s capabilities are determined by its underlying data. Companies often need to clean up their data before these agents can be effective. And finally, this needs to be collaborative. These factors are not separated. They will collaborate with other humans and other agents across the organization to drive outcomes.
CIO.com: You mentioned that AI-driven managed services are different from traditional models. How can CIOs go wrong?
Jeff Baker: The difference with what we call Managed Services 2.0 is that it’s AI-first. Focused on business outcomes.
It’s not just about deploying a team to process tickets and meet service levels. It’s about improving business outcomes over time. Clients who use AI properly see efficiency gains of around 20% in the first year and up to 50% over five years.
The difficult part is how to purchase these services. During the RFP process, procurement teams often try to standardize key elements across vendors. But it can undermine the innovation providers are trying to achieve.
CIO.com: Looking three to five years out, what will separate organizations that succeed with AI from those that remain in pilot mode?
Jeff Baker: It comes down to focusing on business outcomes. What are you trying to accomplish with your technology, people, and organization?
And in a sense, you have to stay out of the agent’s way. They think differently than humans. Too many companies try to treat agent systems like traditional business process automation practices.
Instead, you need to focus these agents on results and allow them to behave as designed. There will be an even bigger impact.
Click here to learn more about PwC Managed Services.
