Conclusion Blog: Signals That Matter

When I started this project back in March, I wasn’t sure exactly what I was getting into. I knew I was interested in venture capital and AI, but I didn’t have a clear sense of what ten weeks of real research would actually look like.

My final presentation went really well. I felt confident delivering it, and the audience found the material interesting — which meant a lot given that this was a non-finance crowd. Getting people who aren’t in venture capital to care about due diligence frameworks and AI evaluation criteria is its own challenge, and I’m glad it landed the way it did.

The core conclusion I reached through this research is something I didn’t fully appreciate when I started: AI doesn’t replace the fundamentals of due diligence, it adds an entirely new layer on top of them. Investors still care about the same things they always have: the team, the market, the business model, the product. But now they also have to ask whether the product is genuinely AI-native or just an AI wrapper, whether the data is handled responsibly, and whether the company can survive in a market where foundation models like ChatGPT are constantly expanding. Those are brand new questions that didn’t exist five years ago, and they sit on top of everything else rather than replacing it.

This project pushed me in ways I didn’t expect. Building frameworks, reaching out to investors cold on LinkedIn, refining a survey, doing a dry run at an actual VC firm all helped me realize the importance of my project. I’m walking away with a much deeper understanding of venture capital and an appreciation for how much rigor goes into evaluating a single startup. I’m grateful to everyone at Bright Ventures, especially Lenore and Ciera, for making this experience what it was.