The Gui Costin Blog

Finding the AI Sweet Spot in Investment Intelligence

Written by Gui Costin | Aug 14, 2025 2:30:00 PM

(Thank you to Balaji.com whose recent post on the Laffer Curve and AI inspired me using his language (slop) to articulate how we have always thought about data, but now from the AI angle.

I give Balaji all the credit for my thoughts and comments below because his articulation hits the nail on the head of how we have always thought about building an information database.)

We’re living through the fastest technology shift of our careers. Artificial Intelligence isn’t some far-off concept anymore. 

It’s here, in our inboxes, our workflows, and, for many of us, in our daily decision-making.

If you’re in the investment business, whether you’re raising capital or putting it to work, AI is changing how you get things done.

But here’s the thing: the right amount of AI is not 100%.

Run with 0% AI and you’ll be too slow to keep pace with today’s markets. Push it to 100%, removing the human entirely, and you end up with what Balaji calls “slop” – fast but flawed results that miss the nuance, context, and accuracy you need. This is especially true with datasets that only rely on unverified data via web scraping.

At Dakota, we’ve found the sweet spot. The magic happens when AI’s speed meets human judgment and verification. Curation and orchestration is the magic for our customers.

Where We Fit in an AI-Driven World

Dakota Marketplace was built for this hybrid reality. We give our customers the clean, structured data they need to make AI work for them, not against them.

Our database coverage includes:

  • 150,000+ LP investor accounts used daily by investment salespeople to raise capital
  • 155,000+ private companies and 12,500+ public companies
  • 5,000+ private company transactions

And every bit of it is human-verified and AI-ready, built for speed, scale, and trust.

When customers feed Dakota’s clean, structured intelligence into their AI tools, or use it via Dakota Marketplaces internet interface, they get results that are both faster and more reliable. 

That’s the sweet spot.

Why This Matters for Our Customers

In fundraising, deal sourcing, and market mapping, bad data is expensive.

It can waste weeks on the wrong prospect (for example bad emails where the analyst has left the firm but you don’t know it). It can erode credibility with an LP.

It can cost an opportunity.

Our customers, whether they’re private equity deal teams, venture capital investors, or investment salespeople, are operating in a world where small teams can now compete with larger ones thanks to AI. 

But to compete effectively, they need the right inputs. That’s where Dakota comes in:

  • We accelerate the “middle” of the process with AI tools by getting accurate information faster;
  • We protect the “edges” (prompting and verifying) with curated, trustworthy data;
  • We give small teams the same research firepower as large firms without the risk of bad data slowing them down.

Ask Dakota Advantage

When we say “human-verified,” we mean more than our in-house team. Through Ask Dakota, we’ve built an always-on feedback loop with our 5,000+ member community of fundraisers, deal teams, and investment professionals.

They see the same data our customers do. If they spot a change, update, or insight, they can flag it instantly. Our team reviews, verifies, and updates Dakota Marketplace in minutes making it the only platform that combines:

  • AI-powered speed
  • Human research verification
  • Community-sourced intelligence

Looking Ahead

AI is only getting faster and more deeply embedded in our workflows. The market will split into two camps:

  1. Teams using AI on top of unreliable data… garbage in, garbage out
  2. Teams using AI with a foundation of clean, verified intelligence. Those are the winners

Dakota Marketplace exists to keep our customers in that second group. For us, AI isn’t about replacing human intelligence. It’s about amplifying it, provided you start with the right data.

To see this in action, book a demo of Dakota Marketplace!