Business intelligence firm Domo launched an AI agent builder alongside an expanded toolkit and library of enterprise data connectors, targeting organizations struggling to deploy custom AI agents across their systems. CEO Josh James framed the release around a core insight: "AI doesn't become valuable when a model gets smarter. It becomes valuable when it's connected." The platform promises to let enterprises build AI agents that can access and act on their existing data infrastructure without extensive technical overhead.
This move highlights the growing recognition that enterprise AI adoption isn't bottlenecked by model capabilities—it's bottlenecked by data access and integration complexity. While startups chase the latest foundation models, established players like Domo are betting that the real value lies in solving the unglamorous plumbing problems. Their approach acknowledges what many enterprise developers already know: connecting AI to legacy systems, managing data permissions, and ensuring reliable data flows matter more than having GPT-5.
What's notable is Domo's positioning as a BI company moving into AI infrastructure rather than an AI-first startup trying to understand enterprise needs. This inside-out approach could prove more practical than the typical AI vendor strategy of building impressive demos that break when they hit real corporate data governance requirements. However, without seeing the actual connector implementations or understanding their approach to data security and compliance, it's hard to evaluate whether this solves real problems or just adds another layer to an already complex stack.
For developers building enterprise AI, Domo's launch signals that data connectivity tooling is becoming a competitive battleground. If you're evaluating AI platforms, focus less on model variety and more on how they handle your specific data sources, permission models, and existing workflows. The winners in enterprise AI won't be those with the fanciest models—they'll be those who make data integration actually work.