July 16, 2020

Backed by Tech Companies, a New Funder Pursues AI Solutions for COVID-19

by Tamara Straus in Article

Trends in tech are also trends in society. That’s as true in philanthropy as it is in business, marketing, science, healthcare, education, you name it. One of the buzzwords we’re hearing increasingly often lately is “digital transformation,” which means pretty much what you think: using digital technology to develop new (and, one hopes, better) ways to sell stuff, conduct research, play, socialize, and so on. To paraphrase a nonprofit tech guy we spoke with recently, COVID-19 is forcing digital transformation across the nonprofit sector right now, ready or not.

Which brings us to the first round of grants out of the C3.ai Digital Transformation Institute, a research consortium established in March, during the early days of the COVID-19 pandemic. The institute was the brainchild of Thomas Siebel, the billionaire CEO of C3.ai, a company that makes software for artificial intelligence applications. In the early weeks of the pandemic, Siebel wanted academics and other researchers to boost their use of AI to address COVID-19, but found that the tech infrastructure was fragmented and unstandardized, hindering progress.

Central to AI computing solutions is the ability to work effectively with large sets of data. Yes, we’ve been creating piles of data for decades, but the mere existence of this virtual infinitude of information doesn’t mean researchers can access or use it. “Tom (Siebel) realized that researchers were unable to work on decent datasets, and that because they didn’t have access to the same kinds of tools, it was hard to compare results and research,” said S. Shankar Sastry, co-director of the C3.ai Digital Transformation Institute. Sastry is also dean of engineering at University of California, Berkeley, one of the institute’s academic research partners.

Siebel wanted to give researchers the software tools and access to handle larger and better datasets that would enable them to study pandemic-related questions, and also to collaborate and share their work. To form the C3.ai Digital Transformation Institute, Siebel pulled in Microsoft and several tech-forward universities and research centers. Its initial mission is to apply artificial intelligence to study and solve the problems of COVID-19 and future pandemics, but going forward, will be supporting other areas of research related to AI.

The C3.ai DTI research partners include University of Illinois at Urbana-Champaign (Siebel’s alma mater), UC Berkeley, Princeton University, University of Chicago, Massachusetts Institute of Technology, Carnegie Mellon University, Stanford University, National Center for Supercomputing Applications, and Lawrence Berkeley National Laboratory. The consortium is governed by Siebel, Sastry and representatives from Microsoft and the University of Illinois, along with an executive advisory committee of academics. The effort is funded by $367 million in grants from C3.ai and Microsoft.

This isn’t Siebel’s first foray into academically oriented philanthropy. Since 1996, the Thomas and Stacey Siebel Foundation has supported scientific research at targeted universities, in topics including stem cell research, energy, computer science and business.

The first round of grants from the new C3.ai Digital Transformation Institute awarded $5.4 million across 26 research proposals, as well as access to C3’s artificial intelligence software and Microsoft’s Azure cloud computing services.

One of the research grantees is Karen Chapple, chair and professor of city and regional planning at UC Berkeley. Chapple’s interdisciplinary project will track housing evictions during and after the COVID-19 outbreak to better understand the risk of housing precarity and inequality, and eventually help public housing policymaking. “Housing is a very tenuous situation, and then you layer on that a health crisis,” said Chapple. Her project will synthesize data about building types, the labor market and income “to understand who is going to be able to pay the rent and who isn’t.”

Such analyses and projections may support better government policymaking. “A lot of folks don’t believe they have a rental crisis in their district until you show them the maps,” Chapple said. “My sense is, this is a very democratic virus—it won’t just be big cities with homeless problems, but will go to many cities with low renter protections.”

Other funded COVID-19 projects include efforts to use AI to assess the risk of serious illness in suspected patients; using genomic data to understand the identification and transmission of the disease; developing models to assist with contact tracing and reopening decisions; studying potential treatments; and vaccine development. The complete list of funded projects is here.

Technology infrastructure and digital transformation are the types of things that only a few outfits in philanthropy have prioritized. Inside Philanthropy recently wrote about some $8.8 million in grants from the Chan Zuckerberg Initiative to support open source software for scientists. Open and more standardized software can improve efficiency of individual research projects, and critically, enable collaboration, transparency and reproducibility—all especially important elements these days, when scientists around the world are focused on aspects of an urgent pandemic problem. Similar issues apply in AI technology and research.

AI has for years been a hot-button issue. Voices from the late physicist Stephen Hawking to Elon Musk have warned about the risks of making machines too smart for our own good, and several funders have backed efforts to study the ethics of AI and otherwise provide oversight for new developments. Just last week, the Rockefeller Foundation released a report calling for expanded discussions to plan for safe and ethical use of AI. It’s important that such discussions take place, now and as the technology grows and develops. After all, if social media platforms like Facebook can have serious and negative impact on society, along with their positive, benign, or silly uses, then AI’s potential dangers shouldn’t be left to chance.

Sastry says that the C3.ai Digital Transformation Institute will keep questions like data privacy and safety at the forefront. “By putting this (institute) together in the common domain and to have the results open, we wanted to create a confidence that social values are protected,” he said.

Read the full article here.