Kumar Garg is the President at Renaissance Philanthropy, a US-based non-profit that, inside its first two years, catalysed greater than $500 million in philanthropic funding globally for science, know-how, and innovation.
The organisation has launched 22 time-bound, thesis-driven philanthropic funds and programmes addressing international challenges. These embrace superior analysis to deal with local weather emergencies, supporting breakthrough concepts for pupil success, responsibly and quickly scaling geologic hydrogen and offering seed funding to researchers and technologists for bettering social service supply.
Previous to becoming a member of Renaissance Philanthropy, Garg labored with Schmidt Futures, the place he helped design and launch moonshot initiatives in schooling. Earlier than that, he helped set price range and coverage priorities for the Obama Administration as a part of the White Home Workplace of Science and Expertise Coverage.
Kumar holds a BA from Dartmouth Faculty and a regulation diploma from Yale Regulation College.
In an interview with indianexpress.com, Garg speaks concerning the objectives and the construction of Renaissance Philanthropy, the moonshot concepts the organisation funds, the hard-to-solve issues in tech, and the impression they’ve created. Edited excerpts:
Venkatesh Kannaiah: Inform us about your journey to Renaissance Philanthropy.
Kumar Garg: I studied political science and laptop science in school and was curious about authorities, so after graduate college, I ended up serving within the Obama administration. I labored for President Barack Obama’s science advisor and obtained publicity to how science and know-how coverage is developed throughout a variety of areas — house, commercialisation, superior manufacturing and math and science schooling.
Story continues beneath this advert
After that, I went to work for Eric Schmidt of Google and helped construct the Science and Tech Basis, Schmidt Futures. It centered on how philanthropic capital might advance completely different fields of analysis and apply them to the general public good.
Renaissance Philanthropy was primarily a spinout. The core workforce got here from working immediately with Eric Schmidt.
Venkatesh Kannaiah: What was the thought behind Renaissance Philanthropy?
Kumar Garg: The concept got here from recognising that the large challenges like local weather, AI, schooling, and financial mobility are issues we have to handle immediately and never later.
So the query turned: might Renaissance construct high-quality science and know-how programmes — accelerating center college math with AI, creating new vaccine platforms, figuring out new vitality sources like geologic hydrogen — and construction them extra like funding funds, however for philanthropy?
Story continues beneath this advert
The mannequin was to construct technical programmes and philanthropic funds round particular sectors, after which go to donors and say: Moderately than constructing all of this internally, you possibly can take part virtually like a Restricted Companion in a enterprise fund.
If we wish extra capital deployed towards exhausting issues, we’d like extra automobiles that make it simpler for individuals to take part. That was the fundamental thought.
In two years, we’ve helped transfer about half a billion {dollars} — roughly $250 million immediately and one other $250 million by advisory assist.
Venkatesh Kannaiah: How do you design your programmes?
Kumar Garg: Our start line is that philanthropy is just one small a part of a a lot bigger system. So the query for us is: what sort of intervention could make a considerable impression inside three to 5 years, in a manner that the broader discipline can ultimately maintain and construct upon?
Story continues beneath this advert
A whole lot of the way in which we design our programmes comes right down to figuring out actually exhausting issues in science or schooling the place centered R&D and technological innovation might make a significant distinction.
Venkatesh Kannaiah: Inform us about a few of your programmes that are creating an impression.
Kumar Garg: For instance, in center college math, we checked out a 2012 J-PAL examine that confirmed intensive, shut tutoring assist might double the speed of studying.
The problem was price. The examine discovered it price round $4,000 per pupil. That’s costly even in america. We’re taking a look at whether or not we will carry it right down to $500.
Story continues beneath this advert
It’s a five-year programme involving a number of groups, all working in direction of the aim. They’re utilizing AI, combining it with established tutoring science, and integrating completely different instruments and strategies right into a coherent system.
It is extremely troublesome to fund the R&D wanted to make this type of work attainable. These initiatives require making use of rising applied sciences, working a whole bunch of pilot research, and testing completely different implementation fashions.
We even have an initiative centered on early studying. Right here, AI is used for screening and evaluation.
One of many main challenges is that educators don’t know a toddler’s precise studying degree, or whether or not the kid could have undiagnosed studying difficulties reminiscent of dyslexia or speech-related points. These screenings are presently costly and troublesome to manage at scale, which implies many youngsters by no means get assessed correctly. With out that info, it’s exhausting to know what intervention is required.
Story continues beneath this advert
AI techniques are already superb at automated speech recognition for adults, however far fewer individuals have centered on constructing speech recognition techniques that work effectively for youngsters.
One other space is tooling for mathematical analysis utilizing AI. We awarded over $18 million to researchers globally to construct instruments that make it simpler for mathematicians to make use of AI of their work.
We’re already seeing a few of the instruments unfold rapidly inside the analysis group. For instance, Lean is a programming language which permits mathematical proofs to be written in a computationally verifiable kind. Among the early grants we supported are already influencing how mathematicians collaborate and the sorts of instruments they use of their day-to-day work.
There may be the Public Profit Innovation Fund. The concept is to enhance one of many main capabilities of presidency, like delivering advantages and companies to residents. Within the US, estimates counsel that just about a trillion {dollars} in advantages go undelivered due to administrative inefficiencies and system complexity.
Story continues beneath this advert
So we requested whether or not rising applied sciences like AI might enhance how these techniques function. A easy instance is name centres. Many authorities companies battle to reply all incoming calls, even for fundamental questions like confirming whether or not an utility has been obtained or whether or not somebody qualifies for a programme.
Among the grants we supported have led to deployments in US states, the place new instruments now assist with eligibility checks, automate routine assist duties, and observe coverage or code adjustments extra successfully.
We’re notably curious about serving to governments undertake AI in a extra experimental and evidence-driven manner.
Venkatesh Kannaiah: Inform us about your accelerator programme.
Kumar Garg: We run a programme referred to as the Massive If True Science Accelerator. The concept behind it’s that formidable researchers usually don’t obtain a lot teaching on creating a very transformative thought. They’re normally centered on working their labs and making use of for grants.
Story continues beneath this advert
So in our accelerator, we provide 15 weeks of teaching and introduce them to governments, donors, and different potential supporters.
Our broader aim is to extend the ambition of each donors and scientists. We’re presently on our third cohort, with round 45 scientists having gone by the programme up to now.
Venkatesh Kannaiah: Inform us about moonshot concepts that you’re funding or seeking to fund.
Kumar Garg: One, which I already talked about, is literacy work, whether or not we will reduce in half the variety of youngsters scuffling with studying by third grade by higher identification of early studying difficulties utilizing AI.
One other space is the position of hidden well being burdens in studying outcomes. One instance is air high quality in colleges. Analysis exhibits that cleaner air in school rooms has a significant impression on each studying and well being outcomes. Even comparatively easy interventions, like higher air purification techniques, can have very excessive returns.
So we’ve been exploring a programme on constructing cheaper, extra deployable air purification applied sciences and ensuring they’re truly utilized in colleges. We’re exploring that each within the US and globally.
One other space we’re curious about is lead air pollution. Lead publicity can have main impacts on IQ and long-term well being outcomes. One thought we’ve been taking a look at is whether or not we will construct a significantly better blood take a look at for detecting lead publicity. The take a look at broadly used immediately was developed roughly 40 years in the past and isn’t exact.
We’re taking a look at the place AI can remodel scientific analysis itself. We have already got a programme in arithmetic, however we’re curious about different domains as effectively.
One space I discover very compelling is monsoon prediction. The Indian authorities has spoken about this problem, however I believe there’s nonetheless room for a way more centered push round superior prediction fashions.
We’ve spoken with researchers who’ve modelled the social advantages of improved monsoon prediction. We’ve additionally begun conversations with donors and governments about whether or not there may very well be assist for a extra concentrated effort on this space. It’s nonetheless at an exploratory stage, however it’s one thing I’m personally very curious about due to the dimensions of potential impression.
We’ve been doing work round geologic hydrogen — the concept that as a substitute of producing hydrogen by industrial processes, you would extract naturally occurring hydrogen immediately from underground sources. If it proves viable, it might develop into an vital software for decarbonising sectors which can be in any other case very troublesome to transition. The dialog now’s about constructing higher subsurface maps, figuring out pilot alternatives, and understanding the place the useful resource potential could exist.
We’ve got additionally been exploring the thought of potassium-enriched salt. There’s already robust randomised managed trial proof suggesting {that a} vital share of heart problems could also be linked to potassium deficiency. By barely rising the potassium content material in salt — with out altering the style — you might be able to meaningfully enhance population-level well being outcomes.
That may very well be particularly vital in nations with excessive charges of hypertension, together with each India and america.
One in all our programmes seems to be at whether or not it’s attainable to construct a brand new technology of space-based telescopes that generate vastly extra information at a fraction of the price of conventional techniques.
Venkatesh Kannaiah: Inform us about science and tech concepts which you suppose are very exhausting to crack.
Kumar Garg: I believe biology is extraordinarily exhausting. Folks usually say, “If we will simply work out this one factor, then AI will resolve the remaining,” or that after a selected breakthrough occurs, all the things else turns into easy. However what we hold discovering is that the deeper you go, the extra complicated the system seems to be.
Take most cancers, for instance. Globally, we’ve made monumental progress, and that progress is constant. However the extra we be taught, the extra we realise it’s not one illness — it’s 1000’s of various subtypes and organic pathways. So at the same time as advances in biology speed up, it stays a deeply complicated, depraved drawback that may require many good individuals engaged on it from a number of angles.
One other problem is ensuring we take into consideration science and know-how issues by way of bottlenecks. Folks usually assume that fixing one rapid challenge will unlock all the things else. However in follow, innovation techniques are normally constrained in a number of methods without delay.
For instance, we’ve been exploring a programme round medical trials — particularly, the right way to speed up their tempo. Most individuals don’t instantly consider that as a science and know-how problem. They suppose science means inventing the subsequent drug or remedy. But when the medical trial system itself is gradual or inefficient, then all of that innovation will get bottlenecked as a result of new remedies can’t attain the market.
We’ve got a fellow on our workforce primarily based in India who has been engaged on how India might modernise its Section 1 medical trial system. Proper now, the method has develop into more and more gradual, whereas nations like China are transferring a lot quicker. Consequently, many promising concepts and corporations merely go elsewhere as a result of they will’t get trials began effectively.
That’s one of many key classes we attempt to emphasise: when fascinated by troublesome science and know-how issues, it’s important to consider carefully about bottlenecks.
Typically the bottleneck is regulation. Typically it’s a scarcity of expertise. Typically it’s that two completely different scientific fields aren’t speaking successfully with one another. Typically it’s funding. However individuals usually mistake probably the most seen or rapid problem for the one problem.
Venkatesh Kannaiah: Do you’re employed with startups? And if that’s the case, how do you interact with them?
Kumar Garg: So the way in which our mannequin works is that we increase cash philanthropically. The donors who assist us are writing cheques with out anticipating monetary returns. The capital comes into the organisation and is then allotted to particular funds or programmes.
As soon as cash enters a selected programme, the fund chief has broad discretion over the way it ought to be deployed. That might imply issuing a grant, making a present, funding a contract, and even utilizing instruments like loans or mission-related investments.
Venkatesh Kannaiah: How do you interact with governments and native innovation ecosystems?
Kumar Garg: We’ve got numerous authorities partnerships, most of them with nationwide governments. For instance, we associate with nationwide innovation companies like ARIA within the UK, SPRIND in Germany, and we not too long ago signed a partnership settlement with the Cupboard Workplace in Japan.
A giant motive governments work with us is that they need to make their R&D ecosystems extra formidable. They need to establish probably the most formidable researchers of their techniques and assist them do their finest work. A whole lot of our programmes are designed particularly to establish these researchers and coach them, which governments discover worthwhile.
Venkatesh Kannaiah: What are your views on the Indian innovation ecosystem?
Kumar Garg: I believe the Indian innovation ecosystem has many strengths. It has a deep engineering base, large manufacturing capability, and main capabilities in areas like prescribed drugs.
India can also be deeply related to the broader Western science and know-how ecosystem, partly due to English and partly due to the Indian diaspora in locations like america. Between the IIT system and the broader technical ecosystem, there’s already quite a lot of scientific and engineering depth.
I believe the large problem now’s the right way to suppose extra systematically about formidable R&D programmes in India. Usually, you’ll see glorious particular person researchers doing robust work, or establishments with fascinating partnerships and pockets of innovation. However should you ask questions like, “What’s India’s equal of the UK Biobank?” — which means a large-scale, deeply structured, high-quality nationwide analysis dataset that many researchers can construct on — it’s not clear what these massive shared moonshot infrastructures appear like.
The query is: what’s the social infrastructure for designing and launching these moonshots? Who’s doing the early-stage scoping work, figuring out the expertise, convening workshops, creating the primary pilot research, and constructing the preliminary momentum earlier than governments step in at scale?
I believe constructing extra of that tradition round formidable experimentation can be extraordinarily worthwhile in India as effectively.