IBM Research Director Dario Gil joins The Final Round to discuss how tech giants are collaborating to combat the coronavirus pandemic.

Video Transcript

MYLES UDLAND: All right, welcome back to "Yahoo Finance Live." Myles Udland here in New York. Let's quickly go to Jared Blikre for a look at how this market settled down. And, Jared, I think not the most encouraging close we've seen, but there have been a lot of volatile moves in the last 30 minutes of trading in the last few weeks. Today just another chapter in that story.

JARED BLIKRE: Yeah, like we've been saying, the markets are illiquid, and any kind of headline can just jolt it one way or another, and today it was to the downside. Bernie Sanders raising objections to the $2 trillion stimulus package.

So we see the Dow ended up almost 500 points but lost nearly 1,000 in the final minutes of trading. And we can see this is the intraday price action, what happened into the close. But I've been looking at the futures and really not moving that much since then.

And I would expect tomorrow coming in that this will probably be largely forgotten. A lot of tailwinds in the market right now.

Let's just go to one of our heat maps real quick. Here's the NASDAQ 100, and it's kind of evenly split. It was mostly green earlier. In the upper left you see Ross is up 13 and a 1/2%. Wynn up nearly 13%. The airlines doing well today. United up 11%.

If we look at the last two days, it's an even more bullish picture here. Yesterday's were some record gains for a lot of companies. But overall, every day brings its own set of challenges, and we certainly had ours today.

MYLES UDLAND: And there will be more in the future. Jared Blikre, thanks for that.

Let's go now to what's happening on the fight against coronavirus and how some of the tech giants out there are helping-- they're doing their part, rather, to work to stop the spread of this virus. Talking now to Daria Gil. He is the Director of IBM Research. Daria, thanks so much for joining us today. And I guess tell us a little bit about what IBM is doing right now and how you guys are using obviously the vast computing power and various things you guys have at your disposal to help, you know, the federal government try to, I guess, track and then contain this virus. What exactly are you guys doing there, and how does that all fit in?

DARIO GIL: Yeah, well, thank you for inviting me. Yeah, this was a very exciting partnership that we created collaborating with the White House Office of Science and Technology Policy and Michael Kratsios specifically, who is the CTO of the United States, as well as with the Department of Energy.

And we coalesced together to bring and create the COVID-19 High Performance Computing Consortium where we have aggregated an unprecedented amount of supercomputers. We're talking about 16 supercomputers with over 330 petaflops of computing power and counting. And then through these public-private partnerships working with MIT, RPI, working with all the other leading technology players-- Google, Amazon, and others-- we've all come together, and we're going to run this consortium so that we can match this capability that is unique in the world and partner with scientists that are fighting, you know, and trying to do new therapies and, you know, antiviral and vaccine development.

ANDY SERWER: Wait, Dario, can you actually just explain a little bit more about the actual workings of this consortium? I'm not exactly clear on how you're using computing power to address this problem.

DARIO GIL: Yeah, let me give you first an example, and then I can share a little bit how it works. So working with IBM Summit, which is the largest supercomputer in the world-- that is owned and managed by the Oak Ridge National Laboratory. Researchers there recently published work where they were collaborating with the University of Tennessee, and they took 8,000 compounds. And what they did is simulations on those 8,000 compounds to figure out what molecules-- what small molecules could bind to the spike protein that is present in the virus that then infects the other cells. And they were able to down select from 8,000 compounds to 77 small molecules that are most likely to then impact and ultimately could affect how the [INAUDIBLE].

So what a supercomputer is able to do is to accelerate the rate of discovery, and that's why this consortium is so important because we all have to do everything in our power to be the best technology expert to compress the rate at which we can do discovery in the context of the pandemic. So that's a good concrete example of something that would have taken months to be able to do in a couple of days, and that is really the purpose of what we're doing here.

RICK NEWMAN: Dario, it's Rick Newman. I've heard IBM in the past describe how Watson, the supercomputer-- if I'm using the right term-- can help speed medical diagnoses and other things like that. So can you apply that sort of artificial intelligence and the computing power to speed the development of vaccines, or are there just some parts of developing a vaccine that you just simply cannot speed up with technology?

DARIO GIL: Yeah, no, it is absolutely fair that there will be parts that are very hard to speed up. But, you know, as you do the scale out, you'll be able to do, you know, like the effects on human health, and you've got to testing enough of the population.

But the key on the situation that we're confronting here is to, you know, accelerate every possible path that we can do simultaneously. There is definitely a front end to the discovery where these can accelerate, and then there's another aspect that has to do with the human testing that will take its due course.

Now we are seeing, however, like, expedited approvals within the context of FDA and other things in terms of clinical trials. So even that is trying to be compressed, but you still have to develop things safely.

What I would say also though is that since the launch of the High Performance Computing Consortium, one of the things we are witnessing is that it becomes also a magnet to attract talent and expertise from many different disciplines, including AI, and say how can we all join forces and prioritize the work we do here? Some of it has to do with antivirals, treatments, and vaccines, but the other ones can also be with other modeling that has to do with the impact of policy, for example, that we can put forth. How will we model what is the progressive return of people, you know, from-- you know, who are working remotely to going back to the offices and the impact that that can have? So the epidemiology aspect of it and the modeling is also going to be important, and supercomputers can help.

AKIKO FUJITA: Dario, you mentioned, you know, sort of other areas of partnership or other technology that can be utilized to try and help coronavirus. I'm wondering, you know, we've seen in some other countries that have had to deal with this, whether it is China or Italy, some increased trafficking of those who have the virus-- you know, having sort of these smart thermometers so you know where the temperatures are and where the hot spots are forming. Are you starting to see that kind of technology being developed behind the scenes in the US? because we're not seeing it in the forefront just yet.

DARIO GIL: Yeah, well, one I mentioned obviously that we are seeing is to take information, you know, all the way down to the county level. Today, for example, we announced through the weather company-- The Weather Channel that we own in IBM, we have released an application to be able to provide information all the way down to the county level, just very much like we have weather today so that everybody can make trusted information based on known cases.

Now what you're asking is what other complementary data could we add or could be added on top of it that could allow for more sophisticated models and reactions around it. I think the issue that we're going to come from there is that it's not that it's not technologically feasible. I think we're going to confront constraints that have to do with social policy and privacy concerns and how will those be managed, right? And a good example of that has been surfacing on the ideas around using location data and be able to use that as complementary parts of the models.

So what I would answer is technically feasible, absolutely yes. There's a lot of other policy implications of doing that though.

MYLES UDLAND: All right, Dario Gil is the research director-- or a director, rather, at IBM Research. Thanks so much for calling into the program. We'll talk to you soon.

DARIO GIL: Thank you.