

20VC: Is More Compute the Answer to Model Performance | Why OpenAI Abandons Products, The Biggest Opportunities They Have Not Taken & Analysing Their Race for AGI | What Companies, AI Labs and Startups Get Wrong About AI with Ethan Mollick
147 snips Jul 31, 2024
Ethan Mollick, an Associate Professor at the Wharton School and Co-Director of the Generative AI Lab, dives into the pressing issues within AI. He discusses how adding more compute might not yield better models, and why OpenAI may be mishandling consumer product development. Ethan critiques the startup landscape and the challenges of AI adoption, emphasizing the need for user-centered approaches. He also explores the importance of local connections in Venture Capital and the evolving role of AI in education as we approach AGI.
AI Snips
Chapters
Books
Transcript
Episode notes
AI's Jagged Performance
- Current AI models excel in some areas but lag in others, creating a jagged performance profile.
- This jaggedness prevents AI from fully replacing human work, as it cannot consistently perform all tasks at a high level.
Steam Train Analogy
- Ethan Mollick uses the steam train analogy to explain how new technologies spread.
- Skilled artisans adapted the steam engine's power to various machines, driving the Industrial Revolution.
Focus on User Needs
- Silicon Valley focuses on scaling for AGI and neglects user needs, leading to poorly designed products.
- Companies should prioritize user-friendly interfaces and practical applications over scaling for theoretical future scenarios.