Google DeepMind CEO predicts artificial general intelligence within a decade

In London, during a briefing at DeepMind's offices, Demis Hassabis, the CEO of Google DeepMind, discussed the future of artificial intelligence and its potential evolution into artificial general intelligence (AGI). He expressed optimism about the emergence of AGI in the next five to ten years, clarifying that this form of AI would be capable of exhibiting capabilities comparable to or exceeding those of humans. Although he acknowledged the current limitations, stating that existing AI systems are vastly passive compared to humans, he emphasized the urgency for further research to tackle these challenges. The primary obstacle to developing AGI is effectively teaching AI to understand and interpret real-world contexts. Hassabis noted that generalizing AI's planning ideas and behaviors for real-world application is a significant hurdle. He elaborated on their efforts, including work with complex games like Starcraft, where diverse agents either compete or cooperate, showcasing the complexities of developing such sophisticated AI systems. In contrast, other industry leaders have differing opinions on the timeline for achieving AGI, with some estimating it to be more than a decade away. This divergence reflects ongoing debates in the tech community regarding the pace of advancements in artificial intelligence. As the conversation around AGI continues, the coordination of interdisciplinary research and development remains crucial to realize systems that can fundamentally match human capabilities.