Google's Demus Hassabis was tasked with turning AI research into profit.

For an academic credited with some of the most significant advances in AI over the past decade, Hasabis is clear on the task ahead: to bring cutting-edge AI technologies to every part of the Google universe to serve its billions of users. Make it to the corner.

“We're like the engine room of the company,” Hasabis told CNBC, talking about his new integrated AI unit within Google.

Last month, Alphabet CEO Sundar Pichai merged Hasabis' DeepMind with Google Brain, a separate AI team, and tapped Hasabis to lead the group. Now it's up to Hasabis to re-establish Google as a leader in creative AI after the company's rapid rise to OpenAI, which is backed by Microsoft.

Google couldn't have a more important job, especially as new generative AI services give users alternative and creative ways to find information online. The business question is, can a long-time researcher like Hasabis be the one to deliver products that consumers love?

Geoffrey Hinton, known as the “Godfather of AI,” says there's no questioning Hassabis' will to win.

“I don't think I've ever met anyone more competitive,” said Hinton, who suggested Google buy DeepMind nearly a decade ago. “Dames is competitive at the level of people who win gold medals at the Olympics.”

He developed this trait at an early age. Hasabis was a child chess prodigy who at one point was ranked No. 2 in the world. He also participated in the World Series of Poker.

Hasabis says they also have customer experience. At just 17 years old, he sent a hit video game called “Theme Park” in the 1990s. Hasabis said that games at the time had to be fun and easy to navigate to be successful. After the theme park, he created his own video game company, Elixir Studios.

Hasabis would later go on to co-found DeepMind, widely recognized as the world's leading AI research lab, attracting some of the most prominent experts in deep learning. When Google acquired the lab for a reported $500 million in 2014, DeepMind was given a long leash to operate independently.

Under Hassabis, DeepMind was known for developing its technology through games like Breakout and AlphaGo, an AI program that defeated the world's top Go player.

There was a practical reason for the focus on games. “It's completely safe, there are no consequences, but they can still learn from it,” Hasabis told CNBC.

During his career at DeepMind and then Google, Hasabis dominated the field of AI.

Even Tesla CEO Elon Musk told the OpenAI co-founders in 2018 that they would need billions of dollars to compete with Hasabis and Google.

However, Google will lose this lead as rivals like OpenAI and Microsoft bring to market headline-making products like ChatGPT and Copilot.

A person who has worked with Hasabis but did not want to be named said that, for some time, Hasabis may have been more interested in winning academic awards than in launching products people could use. . Nature magazine, one of the world's most influential scientific journals, has featured the mathematicians' work several times over the past decade.

“The accolades were never the end goal,” DeepMind said in a statement to CNBC. “They simply reflect the importance and impact of the research they recognized.”

One of DeepMind's most important products, AlphaFold, was a key piece of technology that used AI to help scientists predict the structure of proteins, a huge part of biology for decades. There is a challenge.

DeepMind open-sourced AlphaFold, essentially giving it away for free.

In 2017, a team of Google researchers, separate from DeepMind, published a breakthrough study on Transformers, a way to better process text used for training AI models.

This jumpstarted the wave of AI innovation that followed, particularly in OpenAI. Transformers ChatGPT has a “T”.

Critics have argued that Google was giving away key products and research, and helping its biggest competitor.

“They had the lead and they were very cautious,” Hinton said. “They've been very cautious about both creative AI images and big language models. And when OpenAI worked with Microsoft and ChatGPT was being used by Microsoft, Google couldn't afford to be more cautious. could.”

Accounts and Google went on the offense.

The Washington Post reported in May 2023 that Google was internally announcing a major departure from previous policy. Employees had to stop sharing their research with the world, only publishing papers after the research was turned into products.

AlphaFold became a huge business opportunity, securing commercial partners such as Eli Lilly and Novartis for drug discovery.

During a TED talk, Hasabis said the ChatGPT moment showed that the public found value in LLMs and was willing to embrace them.

“When we're working on these systems, most of the time you're focusing on the flaws and the things they don't do and the illusions,” he said. “We wanted to improve these things before we released them. But the interesting thing is that despite these flaws, tens of millions of people find them very useful.”

Hasabis said it was time to bring these products “out of the rarefied world of science.”

Now investors are waiting to see if Google can pull off what they see as its most important feat, and turn that bleeding edge science into profit.