I’m the founder of a $4 billion software unicorn, and I’ve spent years helping teams unify their data. I know firsthand that even seemingly minor restrictions can disrupt productivity—and completely undermine AI’s potential. 

Pay close attention to what Salesforce did with Slack. What looks like a routine policy update is actually the opening shot in a battle that will decide whether AI can truly revolutionize business. The data wars are coming fast.

Why restricting data is a shot fired

The promise of AI can only be realized with a complete, unified view of all context. For businesses, that means having 100% access to chats, meetings, projects, and files to boost productivity. Most people haven’t seen that value yet because they haven’t been able to bring all the pieces together. Emerging data restrictions will set them back even further.

The new Slack limits cut off the real-time communication context that helps teams make better decisions and get more done. Salesforce knows this, that’s surely why they did it. And while most business leaders don’t care yet, that will change quickly once hard lessons hit.

The irony is painful: Companies are investing millions in AI while at the same time allowing vendors to wall off the data that AI needs to be effective in the first place. 

The clock is ticking

Most enterprises are already struggling with AI value; recent MIT research revealed that as many as 95% of AI projects fail. Look closer and the reason is clear: data fragmentation. Systems can’t access the information they need to be productive, and vendor restrictions will only accelerate the silos that hold AI back.

Too many companies still underestimate the cost of these barriers because AI hasn’t even scratched the surface on potential value. But soon AI agents will begin managing projects, rebalancing budgets, and coordinating teams, and the gaps in context will be glaring.

What enterprises can do

How can companies avoid getting caught in the crossfire? The instinct might be to chase every shiny new AI tool. But that path leads to sprawl, not intelligence. The better answer is convergence: unifying workflows, data, and systems so AI can operate with full context.

That starts with intentional consolidation: cutting back on fragmented apps so AI can act seamlessly across the organization. It also means diversifying strategically so no single vendor’s policy change can derail your roadmap. And it requires betting on open systems with strong APIs and interoperability, so you remain in control of your data.

Above all, enterprises need to treat vendor lock-in and data restrictions as their greatest strategic risk. This isn’t just an IT problem; it’s a board-level issue.

The next enterprise data war is underway, and the restrictions we’ve seen so far are just the beginning. If business leaders anticipate the next walls, rather than wait to react, they will help AI reach its true potential. Otherwise, enterprise AI will be lost to the data war before it has a chance to win.

The opinions expressed in Fortune.com commentary pieces are solely the views of their authors and do not necessarily reflect the opinions and beliefs of Fortune.