In a recent panel discussion, Cox Automotive President Steve Rowley sat down with EVP and Chief Product Officer Marianne Johnson and David Richardson, vice president of AWS AgentCore, to discuss how agentic AI is transforming the automotive industry.
The discussion began by defining the difference between traditional AI tools and agentic AI. While traditional tools assist, agents are able to operate autonomously once given a goal, the right context, and connected data. As Johnson noted, when data is connected and agents are introduced, “those agents can actually talk to each other and create this outcome that you could never imagine before,” a shift she described as “probably the most exciting time in my career.” For dealers, that capability unlocks new ways to scale operations and reach customers more effectively.
That distinction matters because the real value of agentic AI isn’t situational use; it’s systemic impact. It’s easy to leverage AI situationally, but the 10x opportunity comes from giving agentic AI the right context and data to reimagine entire workflows: how service technicians diagnose repairs, how inventory is priced, and how accounting and finance operate. Agentic systems can operate 24/7, connect datapoints, and even collaborate with one another across platforms, making hyper-personalization executable at scale for dealers.
But none of that value is sustainable without trust. The trust dealers work so hard to build with customers remains paramount. As Richardson shared, “trust is hard won, and easily lost,” and protecting that trust requires agentic systems to be built on secure cloud infrastructure, with robust identity management, layered security, auditing, and strict controls over agent behavior. But trust is not built on technology alone. Agentic AI must also reinforce the human element; handling speed and scale so people can focus on what only humans can do: building relationships, earning trust, and applying judgment where it matters most.
The most important takeaway from the discussion was simple: dealers need to start leveraging AI today. Real learning happens through implementation, not observation, and early adopters will be best positioned to operationalize agentic AI responsibly, securely, and at scale.
