VCs say AI companies need proprietary data to stand out from the pack

According to AI companies around the world, they have raised more than $100 billion in venture capital in 2024. Crunchbase dataThis is an increase of more than 80% compared to 2023. This represents almost a third of the total amount of venture capital invested in 2024. That’s a lot of money going into many artificial intelligence companies.

The AI ​​industry has grown so much in the last two years that it is filled with overlapping companies, startups that are still using AI only in marketing but not in practice, and legitimate AI startups that are firing on all cylinders. Investors have to work hard when it comes to finding startups that have the potential to become category leaders. Where do they even begin?

TechCrunch recently interviewed 20 venture capitalists who supports enterprise startups, what gives an AI startup a moat or what sets it apart from its peers. More than half of respondents said AI startups would benefit from the quality or scarcity of their own data.

Paul Drews, managing partner at Salesforce Ventures, told TechCrunch that it’s very difficult for AI startups to have a moat because things change so quickly. He added that he is looking for startups that combine differentiated data, technical research innovation and compelling user experiences.

Jason Mendel, a venture capitalist at Battery Ventures, agrees that technology moats are shrinking. “I look for companies that have deep data and workflows,” Mendel told TechCrunch. “Access to unique, proprietary data allows companies to offer products better than their competitors, and a consistent workflow or user experience allows them to become the core engagement and analytics system that customers rely on every day.”

Having proprietary or hard-to-access data is becoming increasingly important for companies building vertical solutions. Scott Beachuk, a partner at Norwest Venture Partners, said companies that can leverage their unique data are the startups with the most long-term potential.

Andrew Ferguson, vice president of Databricks Ventures, said having rich customer data and data that creates a feedback loop in an AI system makes it more efficient and can also help startups stand out.

Valeria Kogan, General Director Stopthe startup, which uses computer vision to detect pests and diseases on crops, told TechCrunch that it believes one of the reasons Fermata has been able to gain traction is because its model is trained on both customer data and based on the company’s own research and development data. center. The fact that the company labels the data itself also helps improve the accuracy of the model, Kogan added.

Jonathan Lehr, co-founder and general partner of Work-Bench, added that companies aren’t just dealing with data, but how they can clean it up and put it to work. “As a pure seed fund, we focus the majority of our energy on vertical AI capabilities dealing with business-specific workflows that require deep domain knowledge and where AI primarily contributes to previously unavailable (or very expensive) data and cleaning them. it would take hundreds or thousands of man hours,” Lehr said.

Beyond data, venture capitalists say they are looking for artificial intelligence teams led by strong talent that have tight integrations with other technologies, and companies that have deep understanding of customer workflows.