2026 Fortune Brainstorm Tech Roundtable: Inside the AI Decisions Separating Leaders from Everyone Else

AI is making it easier than ever to build. The harder challenge—and the one that came up repeatedly at 2026 Fortune Brainstorm Tech last month—is deciding what to build, how fast to move, and where AI creates the most value.

Those questions were at the center of "Timing, Trade-Offs, and Speed,” a roundtable moderated by Fortune’s Amanda Gerut and featuring West Monroe Chief AI Officer Bret Greenstein.

The discussion also included:

• Sean Bruich, Senior VP and Chief Technology Officer, Amgen
• Dan Gill, Chief Product Officer, Carvana
• Nizar Trigui, Chief Technology Officer, GXO

Each leader brought a different perspective, but the conversation repeatedly turned to the same thing: AI is making it easier to build, test, and experiment. The real challenge is deciding where to focus, and how to move faster without losing sight of what creates business value. 

Below are excerpts from the conversation, highlighting the panelists' perspectives on speed, prioritization, governance, and the future of AI in the enterprise.

You can also watch the full Q&A roundtable here.

Beyond efficiency gains, where is AI creating the greatest business value today?

Greenstein: Everybody talks about cost savings, but I think they're missing the point. Speed is the point. If you can do something in hours that used to take weeks, you don't just save time; you learn faster. You get feedback faster. You can bring value faster and improve faster. Those benefits compound in ways that are almost impossible to measure. That's why I think speed is becoming such a competitive advantage. The challenge, of course, is that it creates pressure. Everyone is trying to figure out if they're keeping up.

Bruich: In biotech, speed matters because outcomes matter. If AI helps scientists discover therapies faster, improve manufacturing reliability, or recruit patients into clinical trials more efficiently, that's meaningful. But speed isn't valuable on its own. It has to be connected to something that advances the mission. One risk right now is that organizations become distracted by the constant stream of announcements and innovations. Leaders have to stay focused on the outcomes they're trying to achieve.

Gill: At Carvana, speed has always been part of how we compete. What's changing now is that AI is dramatically reducing the cost of experimentation. You can prototype faster. You can build faster. You can test ideas that previously weren't worth pursuing. The opportunity is enormous, but so is the temptation to chase too many things at once.

When AI makes it easier to build almost anything, how do leaders avoid building the wrong things?

Gill: I think prioritization is really all there is. The lesson we keep relearning is that one thing all the way done is much more valuable than five things that are 20% done. AI makes prototyping cheaper. Documentation is cheaper. Code generation is cheaper. Suddenly it's easy to do a lot of things. The hard part is still deciding what matters and getting it all the way across the finish line.

Greenstein: AI is making creation dramatically easier, which is exciting. But it also creates a new leadership challenge: deciding what deserves attention. As the incremental cost of work approaches zero, everybody wants to build something. Experimentation creates learning, and that's valuable. But organizations still need a clear connection between AI initiatives and business priorities. If you ask employees how many AI use cases they're exploring, you'll hear hundreds. If you ask executives what matters most to the business right now, the answer should probably be a handful. The goal isn't to limit innovation. It's to make sure innovation is pointed at the problems that matter most.

Bruich: One of the biggest risks is confusing activity with progress. It's easy to launch pilots, proofs of concept, and experiments. Before long, organizations can find themselves managing dozens of initiatives without a clear path to impact. Avoiding the wrong investments requires discipline. Leaders need to evaluate what's working, stop what's not, and focus resources on the ideas that can scale and deliver measurable value. That's where AI moves from being interesting to being transformative.

» Related Content: 7 Ways to Increase Business Speed

How do organizations help employees and leaders adapt to the AI era?

Greenstein: One of the things we're spending a lot of time on is helping people separate their identity from the activities they perform. A lot of people define themselves by the work they do every day. "I build presentations." "I update spreadsheets." Instead, we're encouraging people to think about outcomes. Maybe you don't build presentations, you help leaders make decisions. Maybe you don't update spreadsheets, you help optimize revenue. AI changes activities. It doesn't necessarily change the outcomes you're responsible for.

Trigui: At the end of the day, our business is people. We have more than 120,000 employees. The challenge isn't adopting the technology. The challenge is taking people through that journey and coming out the other side with them. That's why we spend so much time educating leaders. We have AI office hours with executives and board members because if they're going to lead the transformation, they need to understand what's possible.

» Related Content: Build AI as a Business Capability, Not a Technology Project

How do organizations balance speed with trust, governance, and consistency?

Greenstein: We think about it architecturally. There are some things that should be common across the enterprise—how agents connect to systems, how data is accessed, what frameworks are approved. Then there are areas where teams should have freedom to innovate. The people closest to the work usually know best how the work should change. The challenge is letting them innovate without creating a hundred different architectures and a mountain of tech debt.

Trigui: We have a very similar philosophy. Technology leaders need to provide secure platforms, governance, traceability, and controls. Everything must be auditable. Beyond that, experimentation should be encouraged. The goal isn't to slow innovation. The goal is to create an environment where innovation can scale safely. Sometimes you must slow down in a few places so you can move much faster everywhere else.

» Related Content: Enterprise Architecture: Systems Built for Agents

What will separate organizations that win with AI from those that don’t?

Bruich: There’s always a new model, a new technology, or a new headline competing for attention. But there is nothing slower than an unplanned technology detour. Organizations can spend a lot of time chasing the latest innovation only to discover it doesn’t meaningfully advance their goals. The leaders who stay focused on business value, measurable outcomes, and the problems they’re trying to solve will be better positioned to move quickly and scale what works.

Gill: The cost of solving problems is coming down fast, which creates a lot of opportunity. Teams can experiment, prototype, and test ideas much more quickly than they could just a few years ago. But it also makes it easier to take on too much at once. The organizations that create the most value will be the ones that remain disciplined about prioritization and stay focused on the problems that matter most.

Greenstein: I find AI incredibly empowering. There's a moment people have when they realize they can just do things. They can build something. Learn something. Try something they wouldn't have attempted before. If you can help more people reach that moment, and give them the confidence to act on it, your organization is going to move faster.