Venture capitalists say AI companies need special data to stand out

AI companies worldwide raised more than $100 billion in venture capital in 2024, according to Crunchbase dataup more than 80% compared to 2023. It includes nearly a third of all venture capital dollars invested in 2024. That’s a lot of money being funneled into a lot of AI companies.

The AI ​​industry has exploded so much in the last couple of years that it’s become full of overlapping companies, startups that are still only using AI in marketing, but not in practice, and legitimate AI startups are thriving. Investors have a hard time when it comes to finding startups that have the potential to be leaders in their category. Where do they even start?

TechCrunch recently 20 VCs were surveyed Who support startups that build projects around what gives an AI startup a moat, or what makes it different compared to its peers. More than half of survey respondents said that the thing that would give AI startups an advantage is the quality or scarcity of their proprietary data.

Paul Drews, managing partner at Salesforce Ventures, told TechCrunch that it’s very difficult for AI startups to have a moat because the landscape is changing so quickly. He added that he looks for startups that have a combination of differentiated data, innovation in technical research, and a compelling user experience.

Jason Mendel, an investor at Battery Ventures, agrees that technology moats are narrowing. “I look for companies that have deep data and workflow moats,” Mendel told TechCrunch. “Access to unique, proprietary data enables companies to deliver better products than their competitors, while consistent workflows or user experiences allow them to become the platforms for engagement and intelligence that customers rely on every day.”

Access to private or hard-to-obtain data is becoming increasingly important for companies building vertical solutions. Companies that are able to leverage their unique data are startups with long-term potential, said Scott Peychuk, a partner at Norwest Venture Partners.

Having rich data about customers, data that creates a feedback loop in the AI ​​system, makes it more effective and can help startups stand out as well, said Andrew Ferguson, vice president at Databricks Ventures.

Valeria Kogan, CEO Fermataa startup that uses computer vision to detect pests and diseases on crops, told TechCrunch that it believes one reason Fermata was able to gain traction is because its model was trained on customer data and data from the company’s own research and development center. The fact that the company does all of its data classification internally also helps make a difference when it comes to model accuracy, Kogan added.

Jonathan Lear, co-founder and general partner at Work-Bench, added that it’s not just about the data companies have, but also how they’re able to clean it and put it to work. “As a pure seed fund, we focus most of our energy on vertical AI opportunities that address business workflows, require deep domain expertise and where AI is essentially an enabler for capturing and cleaning data that was previously inaccessible (or get “It was very expensive,” Lear said. “It took hundreds or thousands of hours of work.”

Beyond just data, VCs said they look for AI teams led by strong talent, those that have strong integrations with other technologies, and companies that have a deep understanding of customers’ workflows.