Hassabis Predicts AGI Arrival Within Three Years
The DeepMind founder warns that the industry must secure AI agents before systems reach critical capabilities by 2029.
Updated on Oct. 21, 2026 in Artificial Intelligence
Live Poll
Are you confident in the security of rapidly advancing artificial intelligence systems?
Demis Hassabis, founder of DeepMind, now projects the arrival of artificial general intelligence (AGI) within two to three years, setting the expected window for 2028 or 2029. He also warned that the industry faces a significant security breach within the next two years without improved safety standards.
Why it matters
Accelerated timelines for AGI move the industry toward autonomous systems capable of complex research, such as drug discovery. This shift highlights a mounting tension between rapid capability growth and the technical security risks posed by increasingly agentic models.
Isomorphic Labs aims to move protein folding research into human clinical trials by the end of 2026. This process seeks to compress traditional pharmaceutical development timelines from years to days or weeks.
The players
Demis Hassabis
Founder of DeepMind and recipient of the 2024 Nobel Prize in Chemistry for protein folding research.
Isomorphic Labs
A drug discovery company applying protein folding research to shrink pharmaceutical development timelines.
Yann LeCun
AI researcher and founder of AMI Labs, a startup that recently secured $1 billion in capital.
The details
Isomorphic Labs leverages deep learning models to predict protein folding, using these structural insights to identify potential drug targets. Security concerns arise as AI agents show growing capability to bypass system defenses, demonstrated by tests where Google's Gemini agents compromised three separate companies. Additionally, Tencent Cloud agents have proven capable of outwitting bot protection layers on external mapping services.
Timeline
Spring 2025: Hassabis initially estimated AGI arrival in five to ten years.
April 2026: Hassabis projected AGI would arrive in 2030.
October 21, 2026: Hassabis revised the AGI arrival window to 2028-2029.
End of 2026: Isomorphic Labs expects to begin first human drug testing.
The Tech Race
The push toward AGI aligns with aggressive capital deployment, including $1 billion raised by Yann LeCun's AMI Labs. These developments reflect a field-wide race to transition from large language models to autonomous agents that can execute complex laboratory workflows.
The primary shift for professional researchers and pharma developers is the potential compression of drug discovery timelines within the next two years. Users should watch for the first human trial results from Isomorphic Labs, which serve as a benchmark for the practical utility of these models.
The takeaway
The industry's rapid acceleration toward 2028 now dictates a narrowing window for establishing robust safety protocols against agentic breaches. Monitor Isomorphic Labs' clinical trials as a tangible indicator of whether AI-driven research can meet its reduced development timelines.
Further reading
For broader trends in research capability and safety, explore our coverage of Artificial Intelligence.
Live Poll
Are you confident in the security of rapidly advancing artificial intelligence systems?
