At IBC2026, AI was everywhere — in the announcements, the demos, the hallway conversations. Nearly every vendor I talked to was showing off an AI-enabled tool, an MCP server, or some new AI integration between applications. It’s clear that AI is on track to touch nearly everything media organizations do, from agents orchestrating work at the top of the stack to specialized tools handling tasks like analysis and localization underneath.

But underneath the buzz, a quieter conversation is happening, and it comes down to one concern: if AI is going to be everywhere, how do you stay in control? That splits into two questions that are really two sides of the same coin: can you trust the content moving through your supply chain, and can you trust the systems that are increasingly moving it for you? Audiences want assurance that content still reflects the original story after countless format and spec changes, and new EU AI Act transparency requirements are only raising the stakes on disclosing what’s AI-generated or altered.

That second question matters more every day, as AI tools and agents take on more of the work and increasingly hand tasks off to one another automatically. That’s great for efficiency, but no organization wants its AI tools doing things nobody knows about. In the blog, we talk about how a platform like Rally, because it manages execution across the entire supply chain, becomes the system of record for what actually happened: which tools touched an asset, what they did, and whether a person reviewed or approved the work.

My take: these two problems share the same underlying gap — a lack of visibility into what’s actually happening between ingest and delivery. The organizations that close that gap up front, rather than bolting on visibility after the fact, are the ones that will be able to embrace AI while still standing behind the content it touches.