Sir Demis Hassabis, founder of Isomorphic Labs, a four-year-old drug discovery startup owned by Google parent company Alphabet, said an artificial intelligence-designed drug will enter the trial phase by the end of this year.
He told the Financial Times at the World Economic Forum: "We are looking at oncology, cardiovascular, neurodegenerative diseases, all the major disease areas, and I think by the end of this year we will have the first drug. "forum.
"(Discovering) a drug usually takes an average of five to 10 years. Maybe we can speed that up 10 times, which would be an incredible revolution in human health." In October with colleague John Hassabis, who won the Nobel Prize in Chemistry with biochemist David Baker and John Jumper, said.
Isomorphic was spun out of Google's artificial intelligence research arm, Google DeepMind, in 2021, but remains a wholly-owned subsidiary of its parent company, Alphabet. The startup's potential has attracted big pharmaceutical partners keen to reduce costs and make the expensive drug development process more efficient.
Hassabis previously told the Financial Times that his team was working with Eli Lilly and Novartis on six drug development projects.
Hassabis, who is also the CEO of Google DeepMind, said in a wide-ranging interview that the search giant's prototype artificial intelligence assistant, called Project Astra, could be launched to consumers later this year. He described a near future, within three years, when there will be "billions" of AI agents "negotiating with each other on behalf of suppliers and customers" and said this will require a rethinking of the network itself.
He also called on leading AI developers to be more cautious and coordinated in their race to build general artificial intelligence. He warned that if the technology gets out of control or is repurposed by "bad actors," it could threaten human civilization. . . for harmful purposes”.
Hassabis said that the ultimate goal of Google DeepMind is to create general artificial intelligence, or "a system that can demonstrate all human cognitive abilities." He said true general artificial intelligence has yet to be achieved, despite social media's "hype" about how close it is. That's still five to ten years away.
“If something works and is valuable, people will do it,” Hassabis said. "Now that artificial intelligence has passed this stage, the genie can no longer be put back in the bottle . . . so we must work hard to ensure that it is spread around the world in the safest way possible."