Breaking

Hassabis Says AI Is Overhyped Short Term, Underrated Long Term

June 24, 2026 at 10:23 EDT

Foundation Models
Research & Papers

Google DeepMind CEO Demis Hassabis said in a Bloomberg TV interview that AI tends to be overhyped in the near term but is underappreciated over the next decade as it transforms science, health and human creativity.

Demis Hassabis · Google DeepMind

"AI is overhyped in the near term — and underappreciated over the next decade"

The Nobel-winning scientist tempers today's generative-AI frenzy while forecasting a "new golden era of discovery" in medicine, energy and space within 10–15 years.

10×

bigger than the Industrial Revolution — at 10× the speed

100×

potential acceleration of research tasks

~2030

when AGI may arrive (≈3–10 yrs)

The scale he describes

Industrial Revolution vs. the coming AI transformation — same unit, drawn to scale

1×

Industrial
Revolution

10×

Coming AI
transformation

Ten times the magnitude — and arriving ten times faster.

A timeline of impact

Near term · years

No silver-bullet transformation — expectations are excessive.

→

AGI · ≈2030

Roughly 3–10 years away — still some years off.

→

Long term · 10–15 yrs

A golden era of discovery in medicine, energy and space.

The proof it can work

AlphaFold predicted protein structures beyond biologists' expectations — and won the 2024 Nobel Prize in Chemistry. Cited as evidence long-term transformation is plausible.

The limits today

Current models still lack creative and inventive capability — and can't keep learning once training ends. Startup valuations show signs of a bubble.

Continue reading

The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.