Broaden and Build: Finding the Real Value in the AI Revolution

We turned the camera on Operating Partner David Hewit to get his unvarnished take on the state of Venture Capital. His answer was reflexive: Artificial Intelligence.

David brings a sharper lens than most to this conversation. With over 15 years of experience at top-tier asset management firms, including Duquesne Capital and tech-specialist fund Shannon River, he has spent his career analysing technology cycles and capital markets dynamics.

However, David notes that beneath the hype and the staggering productivity gains lies a structural problem, a concentration of capital that is leaving the vast majority of the market on the sidelines.

Here is why David believes it is time to rethink the model, move past the obsession with "decacorns," and get back to first principles.

The Concentration Trap

To understand the current state of VC, you have to follow the money. Right now, approximately 60 cents on every dollar in venture capital is flowing into Artificial Intelligence.

However, if you drill down, you find that the vast majority of that capital is being deployed into the same 20 companies. The industry is funneling billions into the foundational layer, the massive Large Language Models (LLMs) and the infrastructure to run them.

"If you're one of the top multi-billion dollar venture firms that can deploy big rounds into these mega-companies, that's great. But the rest of the venture industry is off on the sidelines."

This concentration creates a dangerous "winner-takes-all" dynamic. It shuts out smaller players and ignores innovators who aren't trying to build the next ChatGPT, but are trying to use that intelligence to solve complex, physical-world problems.

Moving "To The Edge"

The solution isn't to walk away from AI; it’s to look where others aren't.

While the giants fight over who owns the "brain" of AI in the cloud, the next wave of value will come from the "hands and eyes" - AI autonomy at the edge.

We need to return to a first-principles basis when it comes to investment in AI. Instead of chasing the hype of the next consumer chatbot, investors should look for companies applying AI to specific, high-value verticals. We are seeing a shift from generalist intelligence to specialized application, taking global intelligence and applying it to local, high-stakes environments.

This is where the real returns will be found:

• Physical Infrastructure: Using AI to monitor and automate energy grids and marine assets.

• Autonomy: Moving beyond digital assistants to autonomous systems that can operate in real-world weather and environmental conditions.

• Specialized Data: Leveraging unique and proprietary datasets that generalist models don't have access to.

Rethinking Success

We need to dispel the myth that for a venture investment to be successful, the company needs to become a $10 billion "Decacorn".

That mindset forces investors to chase the most crowded trades.

There is a lot of money and strong returns to be made by looking for good but comparitively more modest outcomes companies that build real utility, have a disciplined capital path, and solve tangible problems.

By broadening our scope beyond the "Big 20," we can build a more resilient ecosystem. We can back the companies that are actually doing the work of integrating AI into the fabric of our industries, rather than just the ones making the headlines.

The AI revolution is real, but the smartest money isn't following the crowd, it's building the infrastructure for what comes next.