Quantcast
Channel: Artificial Intelligence
Viewing all articles
Browse latest Browse all 1160

A top Google computer scientist says AI researchers are coming out in force to tackle the COVID-19 crisis, and it could finally usher in the era of AI-aided healthcare (GOOG, GOOGL)

$
0
0

Deepmind, David Silver

  • COVID-19 has become a turning point that could see long-running predictions about how AI will forever change healthcare become a reality.
  • Pundits have for years said that AI will find drugs and cures faster and create a new world of personalized medicine.
  • COVID-19 has caused all the smartest minds in AI to offer their skills to medical researchers en masse, says prominent AI scientist David Silver of Google's DeepMind subsidiary.
  • Silver recently won the coveted ACM Prize in Computing for his work that teach computers how to learn by using games. He's famous for creating AlphaGo in 2016, which beat the world's champion at the board game Go.
  • Visit Business Insider's homepage for more stories.

Tech pundits have promised for years that artificially intelligent supercomputers are supposed to change healthcare forever, finding drugs and cures faster and coming up with personalized treatments for each patient.

But the people who become AI scientists tend to come from the world of computer science and the people who become medical researchers tend to come from the world of medicine. It's historically been two different skill sets, attracting two different types of people, existing in two separate worlds. 

It turns out that COVID-19 has become a tipping point for bringing that promise into a reality as the world's best AI scientists are rushing en masse to offer their skills to coronavirus medical researchers, famous AI scientist David Silver tells Business Insider.

Silver is a principal researcher at Google-owned DeepMind, best known for creating the AlphaGo AI computer that defeated the world champion of the board game Go back in 2016.

"The right way to go forward is for all kinds of scientists to work together, using their skills in collaboration with other scientists," Silver says. "AI by itself, it's not like there's going to be a magic bullet. In collaboration with other scientists, AI and machine learning may be able to accelerate what other scientists are already doing."

He says that challenge has rallied the best minds in AI to meet the challenge.

"We at places like DeepMind are already putting in lots of resources. There are a lot of people thinking about this. I know there are many top minds trying their hardest to see how they can work with others in practice."

This list of initiatives where AI researchers have reached to COVID-19 researchers are too numerous to name. They include the Allen Institute for AI, founded by late Microsoft cofounder Paul Allen, which is using AI to help researchers keep up with the fast-growing corpus of over 45,000 scholarly articles on COVID-19. It's part of the COVID-19 Open Research Dataset. Tech companies everywhere, including Amazon Web Services, Google, Microsoft, and Nvidia, are donating compute time and resources to COVID-19 researchers. 

One area where AI can help is with predictions, sifting through millions of possible combinations to come up with the most likely scenarios.

For instance, researchers at DeepMind recently predicted what some of the unknown protein structures in COVID-19 could be. While the existence of those proteins is just theoretical, those theories based on hard data. DeepMind released these predictions as a model to be used by medical research scientists. The more that researchers can discover the protein structure, the easier it is to develop tests, drugs and therapies. Besides COVID-19, DeepMind is also working on other healthcare issues like eye disease, too.

Earlier this week Silver became the latest recipient of the coveted ACM Prize in Computing for his work that uses games to teach computers how to learn. He's earned a $250,0o0 prize from an endowment from Infosys that will, assuming everything lines up, be formally bestowed upon him at a banquet in San Francisco in June.

The algorithm he helped develop was famously used for to help reduce power consumption at Google's data centers by automatically reacting to the the data center's needs and adjust the heating/cooling systems automatically (much like it would react to the opponents move and adjust its game-playing strategy automatically). It's also been used for planning the trajectories of space probes for the European Space Agency.

Are you a Google insider with insight to share? Contact Julie Bort via email at jbort@businessinsider.com or on encrypted chat app Signal at (970) 430-6112 (no PR inquiries, please). Open DMs on Twitter @Julie188. 

Join the conversation about this story »

NOW WATCH: Why Pikes Peak is the most dangerous racetrack in America


Viewing all articles
Browse latest Browse all 1160

Trending Articles