During a recent lecture at Cambridge University, Demis Hassabis, a Nobel Prize winner in Chemistry, emphasized the transformative role of AI in drug discovery.

He reiterated DeepMind's mission to develop AI responsibly, aiming to solve complex problems efficiently for the benefit of humanity.

Hassabis highlighted the critical importance of AI safety and the necessity to engage with diverse societal stakeholders as AI technology evolves.

He forecasted that advancements in AI could lead to significant progress in various fields, including climate, agriculture, and health, over the next five to ten years.

In addition, he predicted notable advancements in robotics within the next two to three years, driven by enhanced planning systems and world models.

Hassabis expressed confidence in the capabilities of classical Turing machines to tackle complex problems without the need for quantum computing, citing AlphaFold's success in protein folding as a prime example.

He also posited that classical learning algorithms might reveal fundamental patterns in nature, potentially reshaping our understanding of physics and reality.

To illustrate his vision, he introduced 'Project Astra,' a research prototype designed as a virtual assistant capable of understanding and interacting with the physical world to enhance everyday life.

Furthermore, he discussed the development of innovative AI tools like Veo 2, which generates video from text, and Genie 2, which creates computer games from prompts, showcasing their impressive capabilities.

In conclusion, Hassabis reflected on his journey in AI, expressing his belief that artificial general intelligence (AGI) could serve as a powerful tool to deepen our understanding of the universe and humanity's place within it.