A Scale AI exec said AI foundational labs are like movie studios — they make big franchise investments that quickly become irrelevant

In a recent discussion on the 'a16z Podcast,' Ben Scharfstein, Scale AI's head of product for enterprise applications, shared a provocative perspective on the business model of AI foundational companies. He likened these entities, such as OpenAI and Meta AI, to movie studios that pour significant resources into blockbuster projects that may soon lose relevance. Scharfstein's assertion is that foundational AI companies operate more like content studios than traditional software developers. He explained, "My hottest take is I think the right way to think about foundation labs is that they're like movie studios. They invest a ton of money in blockbusters that have a relatively short time span to pay them back." Scale AI, known primarily for its data annotation services that support major tech clients like Google and Meta, is also a player in the generative AI landscape. Following a substantial $14.3 billion investment from Meta in June, the company continues to innovate within this rapidly evolving field. Drawing a comparison to Marvel Studios, Scharfstein pointed out how significant investments lead to valuable franchises, albeit with fleeting returns. He stated, "Marvel, you're investing in 'The Avengers' and you invest a billion dollars in it and then it maybe pays you back over the next 18 months. It's pretty irrelevant after that." He noted that, similar to Marvel's superhero films, firms like OpenAI are developing franchises based on earlier models. "You're able to turn that franchise maybe into a video game and then maybe it is more like software," he added. However, Scharfstein insists that the current operational model resembles that of a content studio more than that of a traditional software company. Scale AI has built its generative AI solutions using technologies from leading models, including OpenAI's GPT-4 and Meta's Llama 2. Recently, OpenAI reported reaching $10 billion in annual recurring revenue within three years of launching its ChatGPT. Despite its growth, Scale AI has faced challenges, evidenced by a 14% reduction in its workforce, impacting 200 employees in its generative AI division. Interim CEO Jason Droeges acknowledged in an internal communication that the company had scaled its capacity too rapidly, resulting in other divisions being left under-resourced. Scale AI has yet to respond to inquiries regarding this situation.