Skip to content
The Twenty Minute VCThe Twenty Minute VC

Demis Hassabis: Why LLMs Will Not Commoditize & Why We Have Not Hit Scaling Laws

Demis Hassabis is the Co-Founder & CEO of Google DeepMind - working on AGI, responsible for AI breakthroughs such as AlphaGo, the first program to beat the world champion at the game of Go; and AlphaFold, which cracked the 50-year grand challenge of protein structure prediction and was recognised with the 2024 Nobel Prize in Chemistry. Demis is revolutionising drug discovery at Isomorphic Labs. Ultimately, trying to understand the fundamental nature of reality. ----------------------------------------------- Timestamps: 00:00 Intro 01:21 What Actually Counts as AGI & Where Are We Today? 02:58 What Are the Biggest Bottlenecks Holding AI Back Today? 03:48 Have We Hit the Limits of Scaling Laws? 04:40 Where Is AI Ahead of Expectations & What's Still Missing? 05:24 Why Can't AI Systems Learn Continuously Like Humans? 06:10 How Did DeepMind Go from Behind to Leading the Pack? 09:10 Are We Heading Toward Model Commoditization? 09:59 What Does the Future of Open Source Really Look Like? 11:25 What Does a Post LLM World Look Like? 13:03 Can AI Really Fix Drug Discovery? 15:01 What Does "Good" AI Regulation Actually Look Like? 17:31 Who Should Be the Ultimate Arbiter of Truth in an AI World? 18:36 If Demis Had One Shot to Fix AI Safety, What Would He Do? 19:58 Is This Time Different for Jobs or Will History Repeat Itself? 24:06 How Do We Solve the Energy Crisis Created by AI? 25:34 Why Stay in the UK Instead of Moving to Silicon Valley? 27:38 Will Europe Ever Build a Trillion-Dollar Tech Giant? 29:20 Meeting Elon Musk for the First Time? 31:03 What Big Questions About AI Is No One Talking About? 31:42 What Does Demis Want His Legacy to Be? ----------------------------------------------- Subscribe on Spotify: https://open.spotify.com/show/3j2KMcZTtgTNBKwtZBMHvl?si=85bc9196860e4466 Subscribe on Apple Podcasts: https://podcasts.apple.com/us/podcast/the-twenty-minute-vc-20vc-venture-capital-startup/id958230465 Follow Harry Stebbings on X: https://twitter.com/HarryStebbings Follow Demis Hassabis on X: https://twitter.com/demishassabis Follow 20VC on Instagram: https://www.instagram.com/20vchq Follow 20VC on TikTok: https://www.tiktok.com/@20vc_tok Visit our Website: https://www.20vc.com Subscribe to our Newsletter: https://www.thetwentyminutevc.com/contact ----------------------------------------------- #20vc #harrystebbings #demishassabis #googledeepmind #deepmind #google #ai #agi

Demis HassabisguestHarry Stebbingshost
Apr 6, 202632mWatch on YouTube ↗

At a glance

WHAT IT’S REALLY ABOUT

Demis Hassabis on AGI timeline, scaling, safety, and science breakthroughs

  1. Hassabis defines AGI as matching the full set of human cognitive capabilities and estimates a strong chance of reaching it within five years based on compute and algorithmic progress trends.
  2. He argues scaling laws have not “hit a wall,” but returns are naturally less explosive than early generations, with compute remaining the main bottleneck both for training and for running meaningful experiments.
  3. DeepMind’s recent acceleration is attributed to consolidating talent and compute across Google and operating with startup-like focus to build larger frontier systems faster.
  4. Key missing capabilities include continual learning, better memory architectures, long-horizon planning, and improved consistency to reduce today’s “jagged intelligence.”
  5. He advocates international minimum safety standards and independent auditing (akin to an atomic-agency model) to mitigate misuse and ensure increasingly agentic systems remain controllable.

IDEAS WORTH REMEMBERING

5 ideas

AGI is benchmarked to the human mind, not narrow test scores.

Hassabis uses humans as the only proven example of general intelligence, so AGI must exhibit the full range of cognitive capabilities rather than excelling at a subset of tasks.

Compute limits progress twice: scaling models and validating ideas.

Beyond training bigger systems, labs need massive compute to test new algorithmic ideas at realistic scale; otherwise promising concepts often fail when integrated into frontier models.

Scaling returns are moderating, but not exhausted.

He rejects the “plateau” framing: performance gains are no longer near-doubling each generation, yet remain substantial enough that frontier labs still see strong ROI from scaling.

Continual learning is a major unsolved capability gap.

Current models struggle to incorporate new knowledge post-training without degrading prior capabilities; Hassabis points to brain-like “consolidation” (e.g., replay/sleep analogs) as a potential direction.

Long context windows are a brute-force stand-in for real memory.

He expects new architectures for memory—beyond stuffing everything into context—to improve efficiency and reliability, especially for agentic systems operating over time.

WORDS WORTH SAVING

5 quotes

We’ve always defined AGI as basically a system that exhibits all the cognitive capabilities the human mind has.

Demis Hassabis

There’s a very good chance of it being within the next five years.

Demis Hassabis

No, I don’t think so… the returns are still very substantial, although they’re a bit less than they were.

Demis Hassabis

I sometimes call these systems jagged intelligences.

Demis Hassabis

Those labs that have capability to invent new algorithmic ideas are gonna start having bigger advantage… as the last set of ideas… all the juice has been wrung out of them.

Demis Hassabis

Definition and timeline for AGICompute as bottleneck: training and experimentationScaling laws vs plateau narrativesContinual learning and consolidation analogiesMemory systems beyond long context windowsPlanning, consistency, and “jagged intelligence”Model commoditization and advantage via new algorithmsOpen source models: lagging frontier, Gemma strategyAI for drug discovery: Isomorphic Labs and trials pipelineRegulation, auditing, and international coordinationJobs, inequality, and productivity distributionEnergy demand, grid optimization, and fusion/materialsUK/Europe tech ecosystem and growth-stage capitalPhilosophical questions: meaning, purpose, consciousness

High quality AI-generated summary created from speaker-labeled transcript.

Get more out of YouTube videos.

High quality summaries for YouTube videos. Accurate transcripts to search & find moments. Powered by ChatGPT & Claude AI.

Add to Chrome