First interview with Scale AI’s CEO: $14B Meta deal, what’s working in enterprise AI, and what frontier labs are building next | Jason Droege

In this interview, Lenny interviews Jason Droege, the CEO of Scale AI, about the evolution of AI models and the role of data labeling and training. Jason discusses Scale AI's current state, its independence from Meta, and the shift towards expert data labeling involving professionals like doctors and engineers. He shares insights on finding and retaining experts, the importance of reinforcement learning environments, and examples of data used to train AI models, including building websites and navigating Salesforce instances. The conversation also covers the challenges enterprises face in adopting AI, the importance of understanding customer incentives, and lessons from Jason's experiences at Uber Eats and other ventures, emphasizing independent thinking, high standards for new businesses, and the significance of gross margins.