Technologies like cloud and Artificial Intelligence (AI) can play a key role in bridging the digital divide in developing economies like India, feels Sriram Rajamani, Distinguished Scientist and Managing Director, Microsoft Research India. Unlike digital revolutions of the past, where the need for specific hardware and literacy requirements raised access barriers, digital advancements powered by cloud and AI can ensure no one gets left behind.
“Previously, technology interaction was driven by computers and the keyboard. Today, advances in machine learning and AI are enabling users to have natural interfaces. Computers now interact with users and take actions through conversational interactions with the user, enabled by Natural Language Processing (NLP) and image processing. These advancements are empowering large number of non-computer users to benefit from the digital revolution,” he says.
“I believe cloud and AI have immense potential to transform society. For instance, the informal work force does not use computers or desktops. However, they use smartphones extensively. Consequently, they are participating in the digital revolution through interfaces such as voice, and through apps. All of this is possible because of cloud and AI. These technologies enable a large portion of the informal workforce to bridge the digital divide. They are also enabling members of the illiterate and semi-literate populace to access and participate in the digital revolution ”
Rajamani has witnessed the impact these technologies can have on societies at large, with Microsoft Research India labs projects like 99DOTS among others, which have been spun off into standalone entities. “Although these technological advances are inspired by India, the potential for impact is not limited to India. Any advances that we make in this direction would be carried on to multiple geographies with similar issues,” he says.
In order to create impact at a larger scale, Microsoft Research India has started the Societal Impact through Cloud and AI (SCAI) program. We caught up with Rajamani to understand the role cloud and AI can play in socio-economic development and how SCAI works. Here are some edited excerpts from our conversation:
What is the mission of Microsoft Research India? What are the areas of research you’re focused on?
Globally, the mission of Microsoft Research is three-fold. Firstly, we work to expand the state of the art in every area in which we do research. Secondly, we help Microsoft with the impact of our research. Thirdly, we contribute deeply to the future vibrancy of the entire computing ecosystem.. In addition, at Microsoft Research India, our goal is to help improve critical mass in the computing research ecosystem in India, as well as design technologies for socio-economic development, inspired by the uniqueness of India’s ecosystem.
We have four focus areas of research in India. The first is foundational work pertaining to algorithms and mathematical foundations of data sciences. Our second area of focus is Artificial Intelligence (AI) and Machine Learning (ML). The third area of interest to us is systems—these include distributed systems security, privacy, programming languages networking, and the infrastructure needs of modern-day computer systems. The fourth area is the role of technology and its impact on the socio-economic development of developing economies.
What’s the role that technologies like cloud and AI can play in developing economies?
Technologies like cloud and AI have a key role to play in bridging the digital divide in developing economies. Today, for example, an increasing number of users in India are accessing the internet because of the affordability factor. Technologies such as AI-powered speech processing are enabling more people to interact with each other. Science and research can play a role in using the cloud and AI infrastructure to drive inclusion. These advancements should be used to benefit the larger population–improve their health, education, job prospects and their livelihood opportunities.
What is the role Microsoft Research is playing in democratizing the benefits of technology?
Since its inception, Microsoft Research India has taken on projects that use technology for socio-economic development. We also have a group called Technologies for Emerging Markets (TEM) which is a world leader in this area. Multiple research projects have graduated from the lab into large scale initiatives.
BlendNet technology enables bulk content like videos to be disseminated through a combination of cloud-enabled metadata systems. The cloud is used to identify a known source of video file nearby. The file is then transferred onto the recipient’s mobile using peer-to-peer local Wi-Fi. We have been working with a non-government organization (NGO) called CGNet Swara in Chhattisgarh, India. The organization runs a voice-based online portal that allows people in the forests of central tribal India to report local news by making a phone call. Technologies such as BlendNet can help the organization to disseminate information where connectivity is poor.
Another example is that of 99DOTS, which is a tuberculosis drug adherence system incubated at Microsoft Research. The success of the project led to its graduation into a separate entity, Everwell, which works with the Indian government and the Gates Foundation. The move enabled the project to achieve the scale required for wide societal impact. Today, Everwell builds infrastructure to help with the Prime Minister’s Tuberculosis Eradication Program in India.
Digital Green is another example. The project was started to disseminate agricultural best practices to smallholder farmers and promote networking among them. Today, the project has evolved into a global development organization. Its reach has spread to tens of thousands of villages across the world.
Essentially, if you look at all these examples, we’re empowering the bottom of the pyramid with cloud and AI.
What’s the genesis of the SCAI program?
We recognized that to create impact at a large scale, such projects require more than research and technology. They require building relationships with the government, NGOs, external partners, impact investors and start-ups. Given our expertise in this area, coupled with the technical strength that Microsoft Research and Microsoft bring to the field, we wanted to create an infrastructure whereby research projects mature outside the lab such that they can be deployed at scale for the betterment of society. That is how we started the Societal impact through Cloud and AI (SCAI) program.
Over the last few years, we have seen a significant increase in projects and enterprises from academia and startups that focus on improving societal conditions. SCAI is a programmatic way of identifying and nurturing projects that have a real chance to create a widespread impact on the society. The motivation for SCAI was to graduate such ideas from inside Microsoft Research Labs to have a sustained presence in the real world.
How does the program work?
SCAI engages with academic groups, startups, and NGOs who want to create a real-world impact by solving societal problems. Typically, these are done through public calls for collaboration which facilitate a match between the prospective partner’s expertise and Microsoft’s interest areas of socio-economic challenges. Last year, 180 respondents answered our first call of entries. Of these, we selected four collaborators as the first set of projects under the SCAI umbrella. They were selected through a highly competitive RFP process.
Further, we have also launched an SCAI Fellow program through which the Lab attracts graduate and undergraduate students as well as working professionals to be a part of SCAI to help with the projects. This approach brings more energy to the collaborators, the researchers and to the mission of SCAI itself.
Thirdly, we actively pursue graduating projects into externally funded spin-offs or unearthing business value and finding a home for the project in Microsoft product groups.
From a collaboration perspective, SCAI works in two ways. One, where the project originates from the lab and is already at a reasonable level of maturity but needs collaboration with external partners to scale. One example would be HAMS which uses AI to ensure driver safety. We are collaborating with various government entities who plan to deploy this technology to automate driver licensing. Another example is our work on financial inclusion, where we are collaborating with a startup that is using the ideas to build a system to manage loans given to auto drivers. Yet another example is our work on natural language processing, where we focus on low-resource languages, and specifically tribal languages like Gondi, where so little data is available.
The second type of project is that where we can grow ideas from external entities with the active collaboration of researchers from Microsoft and Microsoft technologies. This is where our staff bring research expertise and the collaborator the reach that can help deploy our research findings at scale and help advance our research agenda. The final goal in both types of collaborations is to maximize societal impact.
How does SCAI identify partners for collaboration? What are the parameters for choosing projects for collaboration under SCAI?
I would say that there are three parameters for us to choose projects under SCAI. Firstly, the project should address a real-world issue. We should be inspired by the definition of the problem. Secondly, the people driving the project should be of high-quality. Thirdly, there should be a meeting of minds between our staff and the project personnel.
SCAI clearly focuses on projects that have a research component to them, regardless of whether the idea originated from within Microsoft Research or not. Although projects under SCAI do receive funding support, matching of interests is a pre-requisite for Microsoft to make such investments.
We also collaborate with Microsoft for Startups to offer collaborators access to people who are clued into the business side of projects. Working with other entities of Microsoft brings in cross discipline and cross organizational energy into our objectives at SCAI.
What support does Microsoft extend to SCAI collaborators?
The most important support we provide collaborators is access to our expertise and resources of Microsoft Research across geographies. In addition, we also provide access to tools and technologies from Microsoft.
Secondly, we are setting up a physical collaborative space where members of SCAI can come together to brainstorm, exchange ideas. This space will be inside the Microsoft Research India premises in Bengaluru. We see having multiple SCAI projects under the same roof as leading to lead to a great degree of collaboration. The idea is to create an energetic, open environment where everyone can learn from and support the others.
Thirdly, we provide funding to our collaborators. And through SCAI Fellows, we also offer support in terms of skilled personnel.