This is an unusually promising moment for early-stage builders.
That’s what Microsoft Chief Technology Officer Kevin Scott told a group of entrepreneurs and technologists last week at South Park Commons in San Francisco. Tools such as coding agents and large-context models are more capable, he said, and the cost of experimentation is lower than ever. Entire categories are still wide open for people who are willing to test ideas quickly.
In a fireside chat with South Park Commons’ Aditya Agarwal, he said there’s currently a rare window for founders who can ignore the noise, follow the real signals and stay grounded through feedback.
Scott, whose career has spanned academia, startups and large-scale engineering leadership, urged the audience of founders to embrace the opportunity embedded within the dynamic AI landscape.
“I know building things is hard. It’s a grind. Just don’t lose sight of how special this moment is,” he said. “I think everybody in this room has the potential to make a massive impact on the world, by just being fearless about how you put this technology at work for other people.”
The discussion highlighted five ways founders can make the most of this moment.
1. Focus on what’s real, and seek frequent feedback
For Scott, tangible realities define the landscape founders must build within, no matter their willpower or optimism. So it’s crucial to recognize the difference between what someone wishes would happen and the shifts that are already underway, he said.
Successful builders, he said, spot the forces that can’t be ignored — technological, economic or behavioral — and align their work with that trajectory.
But those assumptions have to be pressure-tested constantly, he added, emphasizing the need to validate ideas early and often and to prioritize high-quality signals over noise in the current AI landscape. He encouraged founders to identify the smaller, more reliable indicators that come from actual usage and sustained engagement.
2. Take advantage of the ‘capability overhang’ — but be ready for the grind
Today’s AI systems already contain far more capability than most applications draw upon, Scott said. He highlighted this “capability overhang” as a major opportunity for founders willing to do the messy work required to unlock it.
“Some of the things that you need to do to squeeze the capability out of these systems is just ugly-looking plumbing stuff, or grungy product building,” Scott said. “But you’re in a startup, that’s kind of your life. It’s more about the grind.”
3. Open vs. closed isn’t a battle — it’s a toolbox
Debates often frame open-source and proprietary AI models as opposing camps, but Scott suggested a more pragmatic view.
Different tools serve different purposes, he said, and the most effective solutions will likely combine approaches depending on the problem at hand. The choice is practical, not ideological, he said, and what matters is selecting the right building blocks for the right job.
“The category error here is thinking that it’s got to be either/or,” he said.
4. Embrace experimentation and the unusually low cost of trying things
Scott underscored how cheap and fast experimentation has become. Ideas that once required specialized teams or significant computing resources can now be tested quickly with widely available tools and coding agents. This environment rewards founders who explore widely and follow unexpected findings, he said, and who treat experimentation as an ongoing part of building, not a preliminary step.
“Do the experiments. Try things,” he said. “I would really encourage folks to not be precious about the possibility of failure.”
5. Keep the focus on empowering people and building things of real value
Technology only matters when it expands what people can do, Scott said. Even as AI systems improve rapidly, he said the core question stays constant: Does the work make someone’s life or job meaningfully better?
Scott pointed to the importance of choosing problems worth solving, understanding the human context around them and aiming for outcomes that genuinely help others rather than chasing novelty for its own sake. In a moment when AI can amplify both impact and distraction, he said, grounding product decisions in helping people becomes an anchor — and a differentiator.
“What am I doing with these tools that is of service and of value to my fellow human beings?” he said. “There are certain things for sure that are going to change in pretty dramatic ways, but I think everything that’s changing is also presenting a set of super interesting opportunities for people, because there is a positive-sum empowerment mechanism that’s at work here.”
Check out the full conversation between Scott and Agarwal:
