AI Can Build Everything — But Should It? Founders and the Validation Gap

As part of Global Entrepreneurship Week — a global movement reaching over 10 million people across 200 countries each November — this session tackles one of the defining questions of this year's entrepreneurial moment: AI has made it possible to build almost anything. The real skill now isn't building fast, it's building right.

AI has made it possible to build almost anything, almost instantly. That means the real danger for founders isn't building too slowly — it's building without validation: without knowing they're solving a real problem, or what's worth building in the first place. Join John Avery, Director of ATDC (the Advanced Technology Development Center at Georgia Tech), and Ed Rieker, President of Softlinc, for a practical conversation on using AI well. 

This session explores how AI is reshaping the broader entrepreneurial ecosystem, from how founders validate ideas to how ESOs support them, and where the line sits between using AI as genuine leverage and using it to move fast toward the wrong thing.

We'll dig into two core ideas: why entrepreneurs should be actively using AI as a tool for speed, insight, and efficiency, and why that same speed can be dangerous when it's pointed at a problem nobody actually has. AI can help founders build faster and cheaper, but it can't tell polite interest from real demand, and it often encourages founders to keep building on unproven ideas.

Customer discovery will come up as one example of this tension, but the real focus is bigger: how founders and the ecosystems supporting them can use AI well without letting it replace the fundamental work of understanding real problems and real people.