Summary
- Demis Hassabis predicts AGI could emerge in just a few years.
- He advocates for a new U.S. body to assess advanced AI before its release.
- The proposal suggests that pre-launch testing might become mandatory for top-tier systems.
Demis Hassabis has once again forecasted that artificial general intelligence (AGI) will likely be realized before this decade concludes. Unlike previous technological advancements, he believes AGI will be on par with major discoveries like electricity and fire.
In a recent blog post on X, the CEO of Google DeepMind stated that AGI is "probably only a few short years away," emphasizing its potential to transform human society significantly.
“In the future, when we reflect on this period, I believe we will recognize that we were at the beginning of the singularity—nothing less than the start of a new era for humanity,” he expressed.
Hassabis noted that AGI is distinct from earlier technological advancements, such as the internet or mobile devices, as its implications could be even more profound.
“It is much more akin to the discovery of electricity or fire,” he stated. “When you think about it, we’ve essentially discovered a way to make sand think. It’s astonishing.”
Despite his optimistic outlook, Hassabis cautioned that the rapid evolution of AI capabilities is outpacing society's capacity to comprehend and manage associated risks. He highlighted existing cybersecurity threats from current advanced models and noted that future systems may pose biological, nuclear, and other security challenges.
As AI systems become more autonomous and capable of self-enhancement, he stressed the necessity for stronger technical safeguards to ensure human oversight remains intact.
“In the future, we will require robust protections to maintain control over increasingly autonomous, self-improving systems and to address unknown challenges that will emerge over time,” he remarked.
This announcement aligns with a trend among AI experts who have expressed concerns about the imminent arrival of AGI since the public introduction of ChatGPT in 2022. In January 2026, Anthropic CEO Dario Amodei predicted that human-level AI could surface within one to five years, cautioning that governments might be underestimating the rapid pace of AI development. Then, in June, Hassabis reiterated his forecast that AGI would manifest by 2030, warning that society has "not long to prepare.”
To alleviate these concerns, Hassabis suggested establishing a U.S. Frontier AI Standards Body, inspired by the Financial Industry Regulatory Authority (FINRA), which oversees U.S. brokerage firms. This public-private partnership, primarily funded by the AI sector, would be staffed by independent technical experts and representatives from open-source communities to assess advanced AI models.
“The swift advancements we’re witnessing in AI necessitate a new strategy for evaluating frontier AI capabilities that is dynamic, adaptable, and thorough,” he emphasized. “The U.S. is well-positioned, given its economic and technical prowess, to take the initiative in developing such a framework.”
This proposal follows similar calls from other industry leaders for regulatory oversight of advanced AI technologies.
In May 2023, during a U.S. Senate Judiciary Committee hearing, OpenAI CEO Sam Altman advocated for a federal agency to license powerful AI systems and mandate independent safety evaluations. More recently, President Donald Trump signed an executive order establishing a voluntary framework for the review of advanced AI models prior to their release. In the same month, Dario Amodei warned that AI is becoming excessively powerful and that regulations similar to those of the Federal Aviation Administration (FAA) are urgently needed.
Despite the growing push for AI regulation, Hassabis pointed out that there exists a limited timeframe to set common standards before AGI arrives.
“The future is not predetermined; we must utilize this valuable time before AGI emerges to guide this technology for the benefit of all humanity,” he asserted. “Our collective actions now will shape how the next stage of civilization unfolds. By responsibly introducing AGI into society, we can enter a new era of scientific advancement and progress, leading to a promising future of remarkable human growth.”