In this episode, Dr. Patrick Pilarski, co-author of the Alberta Plan alongside Richard Sutton and Michael Bowling, makes the case for a radically different kind of AI: one that never stops learning, builds its own understanding from raw experience, and gets more out of limited compute rather than more data centers. A Canada CIFAR AI Chair (Amii) and professor at the University of Alberta, Pilarski also leads a lab that builds robotic prosthetic limbs that learn to anticipate what their users intend to do before they do it.
In this conversation, Pilarski joins TechSurge to unpack why his co-author Rich Sutton has publicly called LLMs a dead end, how the Alberta Plan’s “temporal uniformity” differs from how today’s AI labs bolt reinforcement learning onto pre-trained models, and why he believes AI safety problems in today’s LLM agents are really reinforcement learning problems the field has studied for decades.
The conversation covers:
✅ Why Richard Sutton argues large language models are a dead end
✅ The four core commitments of the Alberta Plan for AI research
✅ Why “temporal uniformity” rejects the idea of separate training and deployment phases
✅ Patrick’s upcoming move to ExperienceFlow.AI as Head of Edge Superintelligence
✅ Why he doesn’t believe the AI data center build-out locks the industry into one path
✅ How reinforcement learning’s lessons on exploration and reward specification apply to AI agent safety
✅ Inside his lab’s work on AI-powered prosthetic limbs that predict user intent
✅ Supernumerary limbs and the future of human augmentation
Guest Links:
Patrick Pilarski: Canada CIFAR AI Chair and Fellow at Amii, Professor in the Department of Medicine at the University of Alberta, co-author of the Alberta Plan, and co-lead of the BLINC Lab. Incoming Head of Edge Superintelligence at ExperienceFlow.AI.
Further Reading and Resources
The Alberta Plan – Pilarski’s contrarian roadmap for AI, co-authored with Richard Sutton and Michael Bowling, built around continual learning, temporal uniformity, computational efficiency, and multi-agent interaction.
The Bitter Lesson – Richard Sutton’s influential essay, referenced in the conversation, arguing that scaling compute consistently beats hand-crafted human knowledge in AI research.
Reward Is Enough – The paper discussed in the episode, arguing that sufficiently rich reward signals alone can give rise to the full range of abilities associated with intelligence.
AI Agent Safety is a Reinforcement Learning Problem – A paper co-authored by Pilarski, mapping AI agent safety failures onto established reinforcement learning challenges like safe exploration and reward misspecification.
ExperienceFlow.AI – The company Pilarski is joining as Head of Edge Superintelligence, whose Superintelligence Research Laboratory focuses on continual learning and is led scientifically by Richard Sutton.
AMD’s Acquisition of World Labs – Referenced in the episode’s discussion of “world models,” AMD’s $8.2 billion deal to acquire Fei-Fei Li’s company.
Chapters:
00:00 – Welcome to Edmonton: AI, Free Will, and the Rodeo
00:57 – Why AI Researchers Name Papers Better Than Other Scientists
02:37 – Should We Anthropomorphize AI?
08:03 – The Alberta Plan, Explained
11:25 – “LLMs Are a Dead End”: Contrasting the Alberta Plan with Today’s AI Labs
14:41 – Human Babies, Giraffes, and Built-In Knowledge
17:34 – Four Years Later: What the Alberta Plan Got Right
19:57 – Who’s Actually Pursuing the Alberta Plan Today
21:03 – Breaking News: Patrick Joins ExperienceFlow.AI as Head of Edge Superintelligence
21:40 – What “Edge” Computing Really Means
24:00 – Are We Locked Into the Data Center Build-Out?
26:21 – Reinforcement Learning as Hill Climbing: Rewards and Evals
28:38 – AMD, Fei-Fei Li, and the World Models Debate
33:29 – Why AI Safety Is a Reinforcement Learning Problem
37:27 – Inside the Lab: AI-Powered Prosthetic Limbs
44:20 – How Prosthetics Learn Fast Without “Knocking Over Every Glass”
46:43 – Robot Hands With Eyeballs and Edge Intelligence
47:14 – A Breakthrough: Canada’s First Bone-Anchored Prosthetic Surgery
49:40 – Supernumerary Limbs and the Future of Human Augmentation
53:41 – What’s Underfunded in AI Research Right Now
About TechSurge:
TechSurge Podcast shares the latest insights directly from legendary Silicon Valley leaders, daring new founders, and visionary technologists.
🔔Subscribe for weekly conversations into the intersection of technology advancement, market dynamics, and founder journeys.
What is TechSurge: Deep Tech Podcast?
The TechSurge: Deep Tech VC Podcast explores the frontiers of emerging tech, geopolitics, and business, with conversations tailored for entrepreneurs, technologists, and investment professionals. Presented and hosted by the Celesta Capital team. Send feedback and show ideas to techsurge@celesta.vc.
Each discussion delves into the intersection of technology advancement, market dynamics, and the founder journey, offering insights into the vast opportunities and complex challenges ahead. Episode topics include AI, data center transformation, blockchain, cyber security, healthcare innovation, VC investment trends, tips for first-time founders, and more.
Tune in to hear directly from Silicon Valley leaders, daring new founders, and visionary thinkers. Past guests include Intel CEO Lip-Bu Tan, Micron CEO Sanjay Mehrotra, VC investor Vinod Khosla, and executive leaders from OpenAI, Microsoft, Google, and other leading tech companies.
New episodes release every two weeks. Visit techsurgepodcast.com for more details and to sign up for our newsletter and other content!