Synopsis: In this Techstrong,ai Leadership Insights interview, Joe Davis, executive vice president of platform engineering and artificial intelligence (AI) for ServiceNow, explains how AI agent interoperability will evolve to drive the next era of business process workflows.
Davis argues that as model capabilities improved in 2025, especially around reasoning and accuracy, organizations began deploying AI faster than ever. That acceleration has created a familiar enterprise problem in a new form: tool sprawl. Companies already juggle stacks of vendors and applications; now they’re adding agents on top of them, sometimes imagining a future with “10 AI agents per employee.” The result is an urgent need for orchestration, monitoring and clarity on value.
Davis also says customers are demanding that AI solutions integrate cleanly with other vendors’ systems and AI services, because most enterprises will inevitably run multi-vendor AI environments. He points to Model Context Protocol (MCP) as a practical breakthrough that reduces the friction of connecting AI systems to enterprise software, turning what used to be bespoke integration projects into a repeatable, standardized approach. He also highlights A2A, an open-source protocol under the Linux Foundation aimed at enabling AI-to-AI communication, especially in multimodal, less structured exchanges.
They then go beyond connectivity into experience. Davis expects AI interfaces to evolve from basic text outputs into richer, dynamically generated user experiences—think widgets, dashboards and interactive elements that reduce “UI whiplash” across workflows. At the same time, he notes that long-running autonomous agents raise new operational questions around cost control, safety and governance, reinforcing the need for observability, telemetry and “control tower” style oversight.
Ultimately, Davis frames the next era as workflow-first: agents will proliferate, but enterprise advantage will come from standards, interoperability and governance that turn agent activity into measurable business process outcomes.
