Coatue Pt2. Open AI’s Kevin Weil Dives into All Things ChatGPT | BG2 w/ Bill Gurley & Brad Gerstner

Bg2 Pod Bg2 Pod Jun 21, 2025

Audio Brief

Show transcript
This episode covers a conversation with Kevin Weil, CPO of OpenAI, exploring the rapid evolution and immense future potential of artificial intelligence. There are three key takeaways from this discussion. First, current AI models are the least capable we will ever experience, with exponential improvements constantly on the horizon. Second, AI's utility is rapidly expanding beyond novelty to become an indispensable tool for daily problem-solving. Third, the future of AI involves deeply personalized and proactive assistance, seamlessly integrated into our lives. Weil emphasizes that today's AI models are merely the starting point. The technology is on a steep, continuous curve of improvement, meaning current capabilities will quickly seem basic. Users should expect exponential advancements in AI's power and utility in the very near future. The utility of AI is rapidly expanding from simple tasks to complex, real-world applications. Weil shared a personal example of using ChatGPT to interpret his son's medical report, demonstrating its shift into a tool for immediate, practical problem-solving. This increasing integration is reflected in user engagement, moving from occasional to habitual use. Looking ahead, the next level of AI assistance hinges on personalization and proactivity. Future models will possess memory and context about a user's life, including contacts, calendar, and personal data. This deep integration allows AI to become a truly proactive and indispensable partner, evolving collaboratively through iterative deployment and societal feedback. Overall, the AI revolution is just beginning, promising transformative impact on nearly every product, service, and device we use.

Episode Overview

  • Kevin Weil, CPO of OpenAI, discusses the immense growth and future potential of the company and AI technology.
  • The conversation highlights the rapid pace of AI model improvement, emphasizing that current technology is the "worst" we'll ever use from this day forward.
  • They explore the evolution of user engagement with AI, from novelty to an integrated daily utility for solving real-world problems, such as interpreting medical results.
  • The discussion touches on the importance of iterative deployment, personalization, and memory in making AI an indispensable proactive assistant.

Key Concepts

  • Accelerating Improvement: AI models are on a steep curve of improvement. The capabilities of today's models will seem basic in the near future.
  • Expanding Utility: The use cases for AI are rapidly expanding beyond simple tasks like writing to complex, real-world applications in areas like personal health, research, and data synthesis.
  • Engagement as a Metric for Value: The shift from measuring monthly active users (MAU) to weekly (WAU) and eventually daily (DAU) reflects AI's growing integration and value in users' lives.
  • Iterative Deployment: The strategy of releasing AI capabilities early and often allows the technology to evolve with societal feedback, making the integration smoother and more collaborative.
  • Personalization and Proactivity: The future of personal AI involves it having memory and context about a user's life (contacts, calendar, personal data) to become a proactive and deeply integrated assistant.

Quotes

  • At 00:05 - "The model that you use today is the worst AI model that you'll ever use for the rest of your life." - Kevin Weil explains the rapid and continuous improvement of AI models, suggesting that the future will always bring significantly more capable versions.
  • At 01:19 - "I think it's bigger for sure than I realized... we have grown faster than anything I've ever seen." - Kevin Weil reflects on his first year at OpenAI, stating that the company's growth and the overall opportunity in AI have surpassed his already high initial expectations.
  • At 04:30 - "The first thing I do, of course, is take a picture of it, upload it to ChatGPT and say, 'Hey, should I be worried about this?'" - Kevin Weil provides a personal example of using ChatGPT to interpret his son's complex medical report, demonstrating the shift of AI into a tool for immediate, practical, real-world problem-solving.

Takeaways

  • Today's AI is the starting point, not the destination; expect exponential improvements in capability and utility in the coming months and years.
  • The value of AI is increasingly measured by its integration into daily routines to solve tangible problems, driving engagement from occasional to habitual.
  • Personalization is key to unlocking the next level of AI assistance, as models that understand individual context and memory can become proactive partners.
  • As AI becomes more capable, it will reinvent nearly every product, service, and device we use, creating massive opportunities for both startups and incumbents who can adapt quickly.