Google DeepMind Co-Founder and CEO Sir Demis Hassabis has issued a stark warning that Artificial General Intelligence (AGI)—a system capable of exhibiting all the cognitive capabilities of the human brain—could arrive within “a few short years”. In a comprehensive essay titled A Framework for Frontier AI and the Dawning of a New Age, the Nobel Laureate declared that humanity is currently standing in the “foothills of the singularity”. Rather than standard technological leaps like the internet or mobile phone revolutions, Hassabis compared the impending arrival of AGI to the discovery of fire or electricity. While he noted that a safely managed AGI could serve as a massive catalyst for a “golden age” of scientific progress, drug discovery, and clean energy, he warned that the technology is advancing exponentially faster than society’s current capacity to comprehend it.
To mitigate the escalating commercial and geopolitical race that is outpacing safety protocols, Hassabis has proposed the immediate establishment of a U.S.-led “Frontier AI Standards Body”. Modeled loosely after the Financial Industry Regulatory Authority (FINRA), this independent regulatory watchdog would feature a board composed of leading technical minds and open-source representatives under strict government oversight. The framework calls for AI labs to voluntarily submit their highly advanced frontier models for rigorous testing up to 30 days prior to public launch. The body would comprehensively audit these models for agentic or deceptive behaviors, cybersecurity liabilities, and emerging biological or nuclear risks. Hassabis stressed that these evaluations should eventually transition into a mandatory legal prerequisite before any frontier AI is deployed, serving as a global benchmark to ensure humans permanently maintain control over recursively self-improving systems.
