AI News Roundup — March 2020
by Gabriella Runnels and Macon McLean
The AI News Roundup provides you with our take on the coolest and most interesting Artificial Intelligence (AI) news and developments each month. Stay tuned and feel free to comment with any stories you think we missed!
_________________________________________________________________
The coronavirus currently known as COVID-19 has obviously dominated the news cycle for the last month, so we’re bringing you an all-COVID-19 roundup this month. The regular news is already scary, so we’ve tried to focus on how data science and other advanced methods can help us responsibly assess and combat current and future pathogens. All of us here at Opex wish you the best — stay safe!
Faster Than a Speeding Virus
The world’s fastest supercomputer, nicknamed “Summit” and created by IBM, is working on the all-important problem of the moment: fast-tracking the discovery of drugs that can combat COVID-19 and other viruses like it.
Supercomputers allow biology researchers to quickly test different drug cocktails against the virus, simulating how the specific chemical compounds of a candidate drug would interact with COVID-19 in a real-world scenario. Scientists involved with the project say that Summit has already helped them identify almost eighty promising molecules that might serve as the basis of an effective treatment.
Preparing for the Next Pandemic
Sometimes, AI gets too much hype. As the purveyors of a monthly feature dedicated to AI news, we are fully aware of this, and as our readers, it likely doesn’t surprise you, either. So while we should be excited for AI’s contributions to battling this COVID-19 epidemic, we should of course acknowledge that if AI could truly swoop in and save us from coronavirus, we wouldn’t all be cooped up right now after three months of this virus running amok across the globe.
As a silver lining, this particular global health crisis will help us figure out the most important applications for AI for the likely inevitable (but hopefully far-off) next pandemic. Check out the MIT Technology Review’s great post on our next great global health use cases for AI in this great piece.
Quick Clara-fication
With the severely limited number of COVID-19 tests currently available in the U.S., healthcare workers often have to limit testing primarily to the sickest and most at-risk segments of the population. Young, healthy people with mild symptoms are not likely to get access to formal testing — this is where Clara comes in. (No, we’re not talking about Opex’s Clara, who wrote a lovely blog post about the TechGirlz workshop we did with Unilever last fall; hypothesis testing is totally in her wheelhouse, but testing for viruses, not so much.)
This Clara is a bot jointly developed by Microsoft and the CDC that helps people evaluate their alarming symptoms and decide what action(s) to take. The bot allows for a streamlined, AI-powered pre-screening process that allows users to better understand their own condition, freeing up healthcare resources for the most pressing cases.
Forecasting in Times of Uncertainty
Rob Hyndman, the time series/forecasting legend behind a number of your most-used R forecasting packages, recently weighed in to talk about forecasting COVID-19. Like most stuff he says about forecasting, it’s worth a read.
In this post, he breaks down exactly why “[f]orecasting pandemics is harder than many people think… fitting simple models to the available data is pointless, misleading and dangerous.” The emphasis here is on simple — he stresses that good models for epidemics exist and there’s just enough data for them to be useful. Check out the full post here.
Data See, Data Do’s
Last week, we published a post on our top ten favorite data viz resources for tracking and understanding the coronavirus. Maybe now you feel inspired to create your own viz, but it’s important to remember: with great data comes great responsibility.
This is a painful, uncertain time for billions of people around the world, and a misleading or difficult-to-understand data visualization has the potential to cause avoidable stress and confusion for your audience. To prevent such a misstep, check out this great post on Medium about things you should consider before you create a COVID-related data viz.
That’s it for this month! In case you missed it, here’s last month’s roundup with even more cool AI news. Check back in April for more of the most interesting developments in the AI community (from our point of view, of course).
_________________________________________________________________
If you liked this blog post, check out more of our work, follow us on social media (Twitter, LinkedIn, and Facebook), or join us for our free monthly Academy webinars.
