When Y Combinator, a San Francisco-based startup accelerator, announces a request for startup applications, it causes a stir.
Getting picked by YC means several things for a fledgling startup — $150,000 in seed funding; three months of intensive support, advice, and networking opportunities; the chance to present a business plan before an eager crowd of investors; and a vast network of resources to tap once they graduate. The accelerator’s alumni include Airbnb, Dropbox, Quora, Reddit, Twitch, and Instacart.
All of this is why YC’s announcement on Tuesday that it’s looking to make a massive bet on long-shot climate change technologies could be a big deal.
“This isn’t Plan A,” Sam Altman, Y Combinator president, told me on the phone. “Plan A is clean energy. We’ve been interested in, and investing in, clean energy for a long time now. And I’m just no longer confident that’ll be enough.”
And so it has launched Plan B: investing in long-shot research into ways to cheaply remove the carbon we’ve already emitted. Most of their ideas won’t work — but that’s their whole business model. If they’re able to make just one of these speculative technologies happen or move one of them down the cost curve, it could change the fight against climate change.
Y Combinator wants to take carbon out of the atmosphere fast enough to save the planet
The Intergovernmental Panel on Climate Change (IPCC) reported in its most recent assessment that we need to cut carbon emissions drastically and immediately to hold warming to the moderate levels that will prevent the worst impacts.
But it simply isn’t possible to pull that off without removing a lot of the carbon already in the atmosphere. The IPCC’s more optimistic estimate suggests we need to figure out how to manage 100 gigatons of carbon dioxide removal by 2100.
In the uglier scenarios, we’d need to figure out removal of 1,000 gigatons, the equivalent of 20 years of global greenhouse gas emissions. “All pathways to 1.5 degrees [Celsius of warming] involve large volumes of carbon dioxide removal,” Julio Friedmann, a senior research scholar at the Center for Global Energy Policy at Columbia University, told me. “Carbon removal is an essential, non-negotiable part of a sustainable world.”
We already know how to remove CO2 from the atmosphere. But tried-and-true methods like better forest management, planting forests, and keeping carbon in farm soil are exceptionally land-intensive.
Other, more technological solutions are too expensive to solve the whole problem. For example, direct air capture — dredging CO2 out of the air and distilling it into a usable, or buryable, form — might be starting to get as cheap as $94/ton. If we could maintain that level of cost-effectiveness at billions of time the load, which we probably can’t, as these plants have energy costs of their own, that’d mean removing 100 gigatons would cost us $9.4 trillion. Removing 1,000 gigatons would cost us $94 trillion — nearly the whole gross world product for a year.
So technological approaches to carbon removal have to get cheaper. And the mission of scaling up is reflected on every page of Y Combinator’s writeup of what they’re looking for. They’re interested in investing in research that might make our existing options more cost-effective or less land-intensive, and they’re also interested in researching new options that could surpass the existing ones.
Each speculative technology comes with an estimate: How many gigatons of CO2 could this strip from the atmosphere, “if everything worked perfectly”? For instance, one approach called electro-geochemistry — a simple chemical reaction to turn CO2 into bicarbonates we can dump in the ocean — might manage to consume 90 to 900 gigatons, depending on availability of renewable energy sources for the electricity it needs to function.
Another approach highlighted by the report, ocean phytoplankton that consume CO2, needs no energy. If 1 percent of the ocean were dedicated to this project, it might consume 47 gigatons in a year.
In other words, these bets, if they worked, could help solve our climate problems, perhaps for mere billions of dollars.
Unfortunately, they’re all intensely speculative — years, at best, from being ready to deploy, and quite likely to turn out totally impossible. “These are all very difficult ideas,” Altman told me. “Each one is very unlikely to work individually. And they’re in need of years of R&D.”
But, the idea is, the time to start those years of R&D is now. Then, if we get lucky, and one of the ideas works out, policymakers will have some more options and will be able to carefully consider their tradeoffs.
Here are four ideas Y Combinator wants researched further
Y Combinator is interested in anything in the carbon removal space, but their announcement highlighted four approaches to carbon removal that they’d like to see explored more. These are, YC acknowledges, ideas that “straddle the border between very difficult to science fiction.”
1) The first is ocean phytoplankton. Photosynthesis, a process that happens in all plants, absorbs carbon. And there are vast stretches of the ocean with almost no plants doing any photosynthesis because the water is too nutrient-poor to sustain them. If we engineered better phytoplankton — microscopic marine algae — perhaps our improved phytoplankton could survive with less mineral availability.
Alternatively, we could dump minerals overboard for them, which increases the logistical challenges of the project — it’s not easy to disperse minerals at sea — but simplifies the engineering ones. Existing phytoplankton often release the CO2 they captured when they die, but it might be possible to engineer ones that instead produced a stable form of carbon that could sink to the bottom of the ocean and remain there.
2) The second idea takes advantage of a natural process called mineral weathering. Over time, CO2 reacts with common rocks, forming dissolved mineral bicarbonates. Those bicarbonates are the biggest carbon reservoir in the world. About a billion tons of CO2 get removed from the atmosphere by this mechanism every year.
So what if we could make this process faster? An approach called electro-geochemistry does that. The process sucks up CO2 and water and creates hydrogen and bicarbonates. As a bonus, the bicarbonates combat ocean acidification when released in the ocean.
Mineral weathering is energy-intensive, so it’s not on net a carbon win unless we can power it with renewable energy. Powered by renewables, though, it could strip carbon from the atmosphere at enough scale to hold us under 1.5 degrees of warming.
3) Microbes can synthesize CO2 into other compounds for us, but releasing newly designed microbes into the environment might pose all of the risks associated with any invasive species. So a third area that YC is interested in is developing enzyme systems that don’t involve organisms.
Cell-free synthetic biology is a new field, which means we don’t yet know much about what’s possible and what isn’t. Only the simplest self-sustaining enzyme systems exist today, and this would be an enormously complex one. If researchers cracked it, though, we might be able to build bioreactors that turned CO2 into something stable to dispose of or something useful to sell.
4) A fourth proposal is to flood parts of the deserts of the world, turning them into algae beds that can fix atmospheric CO2. This would be the largest infrastructure project ever undertaken, and while in its simplest form it doesn’t require any new science, getting it cost-effective would require many, many breakthroughs in construction and desalination.
Right now, it’s estimated it’d cost more than 50 trillion dollars. It’d take a whole cascade of new discoveries to make this even potentially feasible, and it’s not clear that the countries with these deserts have any interest in hosting an experiment like this one.
Friedmann, the Columbia researcher, found all of these ideas worthy of additional study, though some seem more promising to him than others. He was the most excited about the mineral weathering: “I think that’s an essential part of a new carbon economy.”
On some other proposals, he was skeptical. “I’m less convinced about the ocean phytoplankton or the desert flooding pathways,” said Friedmann, thinking they were less likely to take in carbon at the volumes required.
That said, he still thought it was valuable for someone to be looking into them. “These are certainly untrodden paths. From the perspective of a startup like Y Combinator, they often make money by treading into white space,” said Friedmann. No one has done enough research to know if these could possibly be viable — and now YC will do that research, even though the answer will often be “no, this isn’t viable.”
I also asked Friedmann about the risks. There are likely to be ecosystem effects from adding algae to the oceans or oases to the desert. We’ll be increasing mineral extraction to do electro-geochemistry at scale.
He thought it was important to stay apprised of risks and to act prudently and carefully — but he wanted to make it clear that inaction is neither prudent nor careful. “We’re already running a crazy unintended experiment on the whole global ecosystem, and the consequences of that are bad and getting worse.”
The actions we take will have risks, and we need to take those seriously — but the risks of inaction, too, are mounting.
Y Combinator wants to offer money, expertise, and connections
This isn’t the first time Y Combinator has put out a request for startups. They’ve done so in biotech, education, health care, clean meat, and other high-impact industries.
I asked Altman if they were usually satisfied with the results. He said that some of their top performers, including billion-dollar Ginkgo Bioworks, came out of such announcements.
“One of the most surprising things to me is that the requests for startups work at all,” he said. “It’s not obvious to me that you can say, ‘We want to fund this thing that sounds ridiculous,’ but the power of the internet is that you can do that, and you hear from people who just happen to be working on that crazy idea.”
Some of these out-there ideas have yielded concrete technical advances, though many haven’t. Ginkgo Bioworks has developed new yeasts, new probiotics, and bacteria that can reduce farmers’ reliance on pesticide.
YC’s decision to add cellular agriculture and clean meat to their list of desired startups may have spurred faster development of some meat alternatives. (Hence the complaint from some tech reporters that “It felt like every other startup was trying to make us try vegan chicken nuggets” at YC’s Summer 2018 batch launch.) And, of course, there have been many, many flops.
I asked Altman his advice for a researcher working on carbon-removal technologies who might not have a business plan, prospective partners, or concrete successes yet. He said, “Apply early. Tell us about the technology and we’ll help you figure out how to make a business model.” YC can’t provide the expertise, but they’re trying to position themselves to provide everything else.
What will YC do for applicants? Altman told me it’s an intensely customized experience. “We’ll figure out where you need help and where we can provide it,” he said. They’ll help nonprofits hunt for funders. They’ll help match individual researchers with partners with the expertise to run a business. They’re looking for both for-profits and nonprofits.
He emphasized that it doesn’t have to be ready to go to market, or even close. After all, they’re not trying to invest in things that will probably work — just in things that might make a huge difference if they did.
None of these proposals have great odds of success. They might still be worth a shot.
Taking big bets is how Y Combinator works. They’ve funded more than 1,900 companies. They’ve made the vast majority of their money off just 17 of those, the ones that have achieved valuations over $1 billion. It’s a business model that has worked extraordinarily well for Y Combinator, which now owns a sizable chunk of startups worth more than $100 billion collectively.
Since 2013, they’ve been experimenting with ways to use this unusual business model — and the expertise and human capital it has let them accumulate — for social change. The first nonprofit accepted into Y Combinator’s accelerator program was Watsi, which identifies inexpensive, high-impact medical interventions in poor countries and lets donors fund them.
They’ve also opened a research division to fund “work that requires a very long time horizon, seeks to answer open-ended questions, or develops technology that shouldn’t be owned by any one company.”
Will researching distant possibilities distract us from the options that already work? It’s hard to predict. It’d be a bad thing if people fixated on climate solutions that are still a distant dream at the expense of solutions that actually exist. This might be part of why the first thing Altman said to me was, “This is Plan B,” and why that’s repeated twice in really big font on their landing page for the project. YC has also invested in Plan A — clean energy — and Altman has pushed for a carbon tax.
Their experience with the startup world has made YC comfortable with funding moonshots. If a project is likely to fail but would be enormously impactful in the unlikely event it succeeded, that’s enough reason to try to make it happen. “These ideas press on the limit of what’s possible, and we’re not sure which side of that line they’re on,” their call for applicants says.
But even if the ideas they solicit are likely to fail, there are benefits to figuring out what’s possible. If an approach turns out to have some insurmountable barriers, then the sooner we know about that, the sooner we can transition our focus to other, more tractable approaches.
Technology is just one piece of the puzzle — but potentially a big one
Climate change was always going to need both technological and political solutions. We needed renewable energy options to replace coal and gasoline. We needed the political will to implement them. We need to emit less carbon, and we need political pressure to enact policies to meet that objective.
These approaches don’t change that. If the research goes as well as could be hoped for, the researchers will take it to policymakers, so all the stakeholders — that is, in a democracy at least, the public at large — can weigh in on what technologies we’re willing to deploy.
What technology can do, though, is lighten the load on political solutions. “It’s hard to get the policy to work when things are expensive,” Friedmann told me. “It’s easy to get the policy to work when things are cheap and work well. That’s where YC has an important role.”
Policy and technology together need to bridge the gap between what’s possible and what’s necessary to save the world — and the more technology is able to bring to the table, the better our odds of making it the rest of the way.
Sign up for the Future Perfect newsletter. Twice a week, you’ll get a roundup of ideas and solutions for tackling our biggest challenges: improving public health, decreasing human and animal suffering, easing catastrophic risks, and — to put it simply — getting better at doing good.