Unsupervised Learning

In this engaging discussion, Daniel Kokotajlo, a former OpenAI researcher and AI alignment advocate, teams up with Thomas Larsen, co-author of the AI 2027 report. They dive into the urgent need for safe AI development, exploring the timeline of AI advancements, the growing tensions between the U.S. and China, and the critical risks of misaligned AI. The duo offers compelling predictions for 2027, emphasizes the importance of public awareness, and lays out essential policy recommendations to ensure a safer AI future.

Michelle Pokrass, a key player in launching GPT-4.1 at OpenAI, delves into the intricacies of AI model development and evaluation. She discusses how user feedback fuels improvements and the evolving challenges in AI training. The conversation highlights the shift towards Reinforcement Fine-Tuning for personalized applications and the importance of balancing conversational abilities with problem-solving as we look towards future models. Her insights reflect a deep understanding of AI's potential and the organizational growth at OpenAI.

When Khan Academy launched Khanmigo, Salman Khan thought they might reach 100k users by 2025. Today, they’re at 1.4 million. 🎧 Sal joined us on Unsupervised Learning to talk about AI's role in education— from the vantage point of someone deploying it at scale. As founder of Khan Academy, he’s overseen the rollout of AI tools to over a million teachers and students, giving him a front-row seat to what’s actually working in classrooms. (0:00) Intro(1:17) The Vision for Future Classrooms(4:28) Khan Academy's AI Initiatives(7:06) Proactive AI and Engagement(10:33) Teacher and Student Experiences(18:24) District-Level Adoption and Policy(25:56) Gamification and Engagement(27:36) Evaluating AI Models(31:03) Global Impact and Future Prospects(37:43) Challenges and Innovations in AI Education(44:19) Quickfire With your co-hosts: @jacobeffron - Partner at Redpoint, Former PM Flatiron Health @patrickachase - Partner at Redpoint, Former ML Engineer LinkedIn @ericabrescia - Former COO Github, Founder Bitnami (acq’d by VMWare) @jordan_segall - Partner at Redpoint

Aidan Gomez, CEO of Cohere and co-author of the Transformers paper, shares insights on scaling AI for real-world applications. He discusses the importance of reasoning in custom AI models and how tailored solutions can outperform general models in enterprise settings. Aidan also highlights challenges in data integration and the significance of privacy in AI use-cases across sectors like finance and healthcare. He emphasizes the need for global collaboration and the evolving landscape of AI technologies, all while maintaining an optimistic view on AI's societal impacts.

Scott Raney and Alex Bard, partners at Redpoint, share their sharp insights on the AI investment landscape. They discuss the strategic imperatives for successful AI companies and unveil the critical distinction between enduring firms and fleeting trends. The duo emphasizes the importance of a balance between domain expertise and AI knowledge within founding teams. They also explore the challenges faced by startups, the paradox of first mover advantages, and what truly drives sustainable growth in the rapidly evolving AI sector.

Join Swyx, an AI podcaster and expert in developer tools, along with Alessio Fanelli, an investor backing innovative AI teams. They discuss the surprising shifts in AI over the past year, including the rise of GPT wrappers and open source model adoption challenges. Insights on Google’s strategic momentum, emerging AI applications, and market trends are shared, alongside a candid exploration of over-hyped and under-hyped trends. The duo also delves into the importance of product-market fit and the complex landscape of AI infrastructure and security.

Nikunj Handa, Product Lead at OpenAI, and Steven Heidel, Engineering Lead, dive deep into OpenAI's new tools for building AI agents. They discuss the transformative potential of multi-agent systems for businesses and how these innovations enhance customer interactions. The duo also explores challenges in fine-tuning AI, navigating the evolving infrastructure landscape, and balancing usability with customization in APIs. Their insights highlight a future where AI plays an integral role in automating tasks across various industries.

Noam Shazeer, co-inventor of the Transformer, and Jack Rae, Research Director at DeepMind, dive into the future of AI. They discuss the groundbreaking capabilities of Gemini 2.0 for reasoning and creativity. The duo explores the complexities of AI evaluation metrics and the evolving role of test-time compute, emphasizing efficiency over traditional methods. They also reflect on the philosophical challenges of AGI, the rise of vision-based models, and AI's transformative impact on healthcare and education, highlighting the balance between innovation and safety.

In this discussion, Mike Schroepfer, former CTO of Meta and founder of Gigascale Capital, dives into the crucial link between AI and energy. He reveals how scaling energy production can democratize AI access and tackle climate issues. Mike reflects on his decade at Meta, highlighting AI's transformative role in tech leadership. He also shares insights on future AI tools, the integration of VR, and the impact of open-source development in accelerating innovation. Get ready for a glimpse into the future of AI and its potential to reshape our world!

Vincent Vanhoucke, a Distinguished Engineer at Waymo and former Head of Robotics at DeepMind, dives into the cutting-edge world of autonomous driving. He discusses the integration of large language models in self-driving technology and the challenges of urban deployment. Vanhoucke shares his insights on achieving superhuman driving standards through advanced sensor technologies. He also highlights the evolving role of AI in automation, from reasoning capabilities to innovative applications in industries like non-animal cheese production.