Demis Hassabis: Major AI breakthroughs come from a few key labs, AGI could be achieved in five years, and computational resources are vital for innovation | 20VC

AI breakthroughs could lead to achieving artificial general intelligence within the next five years.

Listen on 20VC

Share

Add us on Google by Editorial Team Apr. 10, 2026

Key takeaways

• Major AI breakthroughs predominantly originate from a few key research labs.
• Algorithmic innovation will be crucial for labs to maintain a competitive edge.
• AGI is defined by its ability to mimic all human cognitive capabilities.
• Achieving AGI within the next five years is a realistic possibility.
• Computational resources are vital for both scaling AI systems and validating new ideas.
• DeepMind is expected to continue leading in AI breakthroughs.
• Current AI systems lack continuous learning capabilities post-training.
• Leading AI labs are pulling ahead due to their innovation capabilities.
• AI systems currently struggle with long-term planning and consistency.
• Open source models trail frontier models by approximately six months.
• The concentration of AI innovation underscores the importance of leading labs.
• Continuous innovation in algorithms is necessary for future AI advancements.
• AGI’s definition emphasizes the brain as the only existing proof of general intelligence.
• The timeline for AGI development reflects significant progress in AI technology.
• Computational power is essential for AI research and development.

Guest intro

Demis Hassabis is the Co-Founder and CEO of Google DeepMind. He led the development of AlphaGo, the first program to beat a world champion at the game of Go, and AlphaFold, which solved the 50-year challenge of protein structure prediction and earned the 2024 Nobel Prize in Chemistry. At Isomorphic Labs, he is revolutionizing drug discovery through AI.

The dominance of key AI research labs

• I would say about 90% of the breakthroughs that underpin the modern AI industry were done by either by Google Brain or Google Research or DeepMind.

  — Demis Hassabis

• The majority of AI breakthroughs come from a few leading labs, indicating a concentration of innovation.
• These labs are pivotal in shaping the future of AI with their groundbreaking research.
• The competitive landscape of AI research is heavily influenced by these key players.
• Innovation in AI is largely driven by the capabilities of these research labs.

• Those labs that have capability to you know invent new algorithmic ideas are gonna start having bigger advantage over the next few years.

  — Demis Hassabis

• The ability to innovate algorithmically will determine the future success of AI labs.
• The concentration of breakthroughs highlights the strategic importance of these labs.

Defining artificial general intelligence (AGI)

• We’ve been very consistent how we define agi as basically a system that exhibits all the cognitive capabilities the human mind has.

  — Demis Hassabis

• AGI is characterized by its ability to replicate human cognitive functions.
• The definition of AGI underscores the brain as the only existing proof of general intelligence.
• Understanding AGI is crucial for discussions about the future of AI.
• The pursuit of AGI involves replicating the cognitive capabilities of the human mind.
• AGI’s definition is pivotal in guiding AI research and development.
• The significance of AGI lies in its potential to mimic human intelligence comprehensively.
• The development of AGI is a major milestone in the field of artificial intelligence.

The timeline for achieving AGI

• I would say there’s a very good chance of it being within the next five years so that’s not long at all.

  — Demis Hassabis

• The possibility of achieving AGI within five years reflects rapid advancements in AI.
• This timeline indicates significant progress in AI technology.
• Achieving AGI represents a major technological milestone.
• The forecast for AGI development highlights the pace of AI innovation.
• The potential for AGI within five years underscores the urgency of AI research.
• The timeline for AGI development reflects expert analysis on AI progress.
• Achieving AGI would mark a transformative moment in the field of artificial intelligence.

The role of computational resources in AI

• You need quite a lot of compute if you have a lot of researchers with lots of new ideas.

  — Demis Hassabis

• Compute is essential for scaling AI systems and conducting experiments.
• Computational resources are critical for validating new AI ideas.
• The dual role of compute is crucial for understanding AI development.
• Access to compute is a key factor in advancing AI research.
• The importance of compute highlights the resource-intensive nature of AI development.
• Computational power is a fundamental requirement for AI innovation.
• The reliance on compute underscores the technical demands of AI research.

DeepMind’s ongoing contributions to AI

• I would back us to sort of make those breakthroughs in the future if there are any missing ones.

  — Demis Hassabis

• DeepMind is expected to continue making significant breakthroughs in AI.
• The lab’s track record reflects confidence in its research capabilities.
• DeepMind’s historical contributions position it as a leader in AI innovation.
• The lab’s ongoing research is pivotal for future AI advancements.
• DeepMind’s role in AI underscores its strategic importance in the field.
• The lab’s contributions highlight its influence on the direction of AI research.
• DeepMind’s breakthroughs are crucial for the evolution of artificial intelligence.

Limitations of current AI systems

• These systems don’t learn after you finish training them… the brain does this very elegantly.

  — Demis Hassabis

• Current AI systems lack the ability to learn continuously after training.
• This limitation suggests a direction for future AI research.
• Continuous learning is a critical aspect of human cognitive capabilities.
• The inability to learn post-training highlights a gap in AI development.
• Addressing this limitation is essential for advancing AI technology.
• The challenge of continuous learning underscores the complexity of AI systems.
• Overcoming this limitation is crucial for achieving true general intelligence.

The competitive edge of leading AI labs

• I feel like maybe you know the three or four leading labs now which we’re one I think the gap is sort of starting to pull away.

  — Demis Hassabis

• Leading AI labs are pulling ahead due to their ability to innovate algorithmically.
• The competitive landscape of AI research is shaped by these labs’ capabilities.
• Innovation in algorithms is a key factor in maintaining a competitive edge.
• The gap between leading labs and others is widening due to innovation.
• The ability to innovate is crucial for future success in AI research.
• Leading labs’ competitive edge highlights the importance of continuous innovation.
• The dynamics of AI research are influenced by the capabilities of these labs.

Challenges in achieving general intelligence

• These systems are not very good at planning at long time horizons… maybe one of the biggest is consistency.

  — Demis Hassabis

• Current AI systems struggle with long-term planning and consistency.
• These challenges are essential for achieving general intelligence.
• Addressing these limitations is crucial for the evolution of AI systems.
• The struggle with long-term planning highlights a gap in AI capabilities.
• Consistency is a critical characteristic needed for true general intelligence.
• Overcoming these challenges is necessary for advancing AI technology.
• The limitations in planning and consistency underscore the complexity of AI development.

The evolution of open source models

• Open source models are probably one step back from the absolute frontier… it usually takes about six months for the open source community to sort of reimplement and figure out what those ideas are.

  — Demis Hassabis

• Open source models will continue to evolve but lag behind frontier models.
• The timeline for open source advancements highlights the role of the community.
• Open source models play a significant role in the AI ecosystem.
• The evolution of open source models reflects the collaborative nature of AI development.
• The lag in open source models underscores the challenges of keeping pace with frontier models.
• The relationship between open source and frontier models is crucial for understanding AI advancements.
• The role of open source models highlights the diversity of approaches in AI research.

Disclosure: This article was edited by Editorial Team. For more information on how we create and review content, see our Editorial Policy.