DeepMind CEO Sees AGI Arriving by 2029-2030, Urges Faster Preparation
Updated
Updated · Computerworld · May 28
DeepMind CEO Sees AGI Arriving by 2029-2030, Urges Faster Preparation
1 articles · Updated · Computerworld · May 28
Summary
Demis Hassabis said artificial general intelligence could arrive around 2030 — possibly in 2029 or sooner — leaving society only a few years to prepare.
Axios quoted the DeepMind chief saying progress is moving faster than expected, with today’s autonomous AI agents serving as a “practice run” for much more powerful systems.
Governments, economists and the broader public are not taking the shift seriously enough, Hassabis warned, despite the need to plan for its social and economic effects.
Leading AI labs are already focused on systems that can help improve their own development, he said, a capability he argued could speed research but also introduce new risks.
With AI now designing its own successors, are we prepared for an intelligence that could soon surpass our own?
As AGI promises to automate R&D, what fundamental hurdles could still derail its arrival by the 2030 deadline?
Can national laws govern AGI when its development and impact will be inherently global and borderless?
DeepMind’s 2029 AGI Timeline: Why Businesses and Governments Must Act Now
Overview
DeepMind CEO Demis Hassabis has updated his forecast, predicting that Artificial General Intelligence (AGI) could arrive as early as 2029. This announcement, made in May 2026, highlights the urgent need for society to prepare for rapid AI advancements. Hassabis warns that businesses and governments are not ready for the speed of progress, especially as unexpected capabilities emerge in new AI systems. He emphasizes both the benefits and risks of AI systems improving themselves, urging immediate action to adapt to the accelerating pace of innovation and to manage the profound changes AGI will bring.