The Pros And Cons Of Open Source AI Business Models

Investors Weigh In on the Different Approaches

The growing popularity of generative AI has given rise to two distinct paths for startups: those opting for closed source models and those embracing open source. Each approach has its own set of advantages and disadvantages, and investors have varying opinions on which is the better choice. Let’s take a closer look at what some investors have to say about the pros and cons of open source AI business models.

Transparency and Trust

Dave Munichiello, a general partner at GV, believes that open source AI innovation fosters trust in customers through transparency. Open source models, methods, and datasets allow developers to assess and probe the technology, creating a sense of confidence in the quality of the model. Munichiello acknowledges that closed source models may be more performant, but their lack of transparency makes it harder to build trust, especially among boards and executives who are looking for endorsements from brand-name tech companies.

Polished User Interfaces

Ganesh Bell, the managing director at Insight Partners, agrees with the advantages of open source AI models, but he highlights one downside: the user interfaces (UI). According to Bell, open source projects often have less polished front ends, making them less consistent and harder to maintain and integrate. While open source AI models excel in transparency, their UI elements may require additional effort to provide a seamless user experience.

Go-to-Market Strategy and Customer Expectations

Christian Noske, a partner at NGP Capital, believes that the choice between closed source and open source AI models matters less for startups than their overall go-to-market strategy. Regardless of whether the model is open source or not, Noske suggests that startups should focus on applying the outputs of their models to business logic and proving a return on investment for their customers. He adds that many customers are not concerned with the underlying model’s source, as they are primarily interested in solving their business problems.

The Impact of Regulation

As the AI industry faces increasing scrutiny, regulation becomes a factor that could shape how startups grow and scale their businesses. Noske sees regulation potentially adding costs to the product development cycle, favoring big tech companies and incumbents over smaller AI vendors. However, he acknowledges the need for clear and responsible use of data in AI, labor market considerations, and regulations to prevent the weaponization of AI. In contrast, Bell views regulation as a potentially lucrative market, as companies building tools and frameworks to help AI vendors comply with regulations could contribute to building trust in AI technologies.