AI companies worldwide will receive more than $100 billion in venture capital by 2024; Crunchbase dataThis is an increase of over 80% compared to 2023. That covers nearly a third of total VC dollars invested by 2024. This is paying a lot of money for many AI companies.
The AI industry has become so vast over the past two years that it's filled with overlapping companies and startups that are still using AI in marketing, but not in practice. AI startups are disappearing from the legal diamond ring. Investors have their work cut out for them when looking for startups with the potential to be category leaders. Where do they start?
TechCrunch recently 20 VCs were surveyed. A backer of startups for businesses on what makes an AI startup viable or what makes it different compared to its peers. More than half of respondents said what would hold AI startups back 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 cash in because the landscape is changing so quickly. integration of different information; He added that they are looking for startups that combine technological research innovation and engaging user experience.
Jason Mendel, a venture capitalist at Battery Ventures, agreed that technology moats are shrinking. “I look for companies with deep data and workflows,” Mendel told TechCrunch. “Access to unique, proprietary data enables companies to deliver better products than their competitors, and a sticky workflow or user experience allows them to become core systems of engagement and intelligence used every day.”
For companies building vertical solutions, having private or hard-to-get data is becoming increasingly important. Scott Beechuk, a partner at Norwest Venture Partners, says companies that can use their unique data in-house are the startups with the most long-term potential.
Andrew Ferguson, vice president at Databricks Ventures, said having rich customer data and data that creates a feedback loop in an AI system can help startups stand out and be more efficient.
CEO Valeria Kogan; StopFermata, a startup that uses computer vision to detect pests and diseases on crops, told TechCrunch that it thinks its model is trained on both customer data and data from the company's own research and development. center. Having the company do all the data labeling in-house makes a difference when it comes to the model's accuracy, Kogan said.
Jonathan Lehr, co-founder and general partner at Work-Bench, added that it's not just about companies' data, but how they can clean it and make it work. “As a Pureplay seed fund, we're focusing most of our energy on vertical AI opportunities that address business workflows that require deep domain expertise, and AI is key to accessing and cleaning data that was previously inaccessible (or too expensive) in a way that would take hundreds and thousands of hours,” Lehr said.
In addition to data, VCs have strong talent leading AI teams; Companies with strong integrations with other technologies; He said he was looking for companies with a deep understanding of customer workflows.