Demis Hassabis: AGI Is Near and May Outpace the Industrial Revolution Tenfold

AGI on the Horizon

Demis Hassabis, Nobel‑winning AI pioneer behind AlphaGo, AlphaFold and the Gemini models, asserts that artificial general intelligence—capable of the full range of human cognition—may be only a few short years away.

The Scale of the Coming Transformation

He likens AGI’s potential impact to a technological leap ten times larger and ten times faster than the Industrial Revolution, suggesting that breakthroughs in drug discovery, clean energy, and advanced materials could occur within months rather than decades, ultimately leading to a world where resources are no longer a limiting factor.

Risks and the Dual‑Edged Competition

Hassabis warns that frontier models already pose security challenges: they can locate system vulnerabilities, generate malicious code, and may soon enable threats in nuclear, bio‑security, and other high‑risk domains. He stresses the need for strong safeguards to keep humanity in control of recursively self‑improving systems.

Call for a Frontier‑AI Standards Body

He proposes a U.S.–led standards organization, modeled after financial regulators, to dynamically test and certify “frontier‑class” models against regularly updated benchmarks. Organizations meeting these thresholds would be designated as frontier labs, required to publish model cards, maintain rigorous internal security, and undergo pre‑release evaluations up to 30 days before deployment.

Evaluation Framework and Governance

The evaluation process should employ strict scientific methods, testing for security, bio‑threat, and other high‑risk capabilities, and include safeguards such as digital watermarks and human‑readable output tokens. Benchmarks would be refreshed quarterly, with an independent test set to prevent over‑fitting, and a third‑party audit ecosystem would be established under government cooperation.

Global Collaboration and Future Outlook

Hassabis envisions this framework as a springboard for international consensus, balancing innovation with responsibility. He emphasizes that solving technical challenges is only the first step; broader economic, philosophical, and societal questions will arise as AGI reshapes humanity’s purpose and prosperity.