Axios Future

New and old machine-learning tools are helping scientists sift through the flood of research produced on COVID-19.

Why it matters: The coronavirus pandemic has led to an unprecedented wave of scientific publications on every aspect of the virus and potential treatments.

• But without better tools to pick out meaningful research, it's too easy to miss the science that matters — or be misled by headline-grabbing papers that turn out to be wrong.

By the numbers: According to the COVID-19 Primer, a public dashboard created by the machine-learning company Primer AI, researchers had published 27,569 papers about COVID-19 as of June 17.

• Of those, 21,000 went through the scientific peer review process, meaning that experts in the field have examined the publications and researchers have more confidence in the results. 6,569 are what are known as preprint papers, meaning they were put out to the public before the peer review process.
• "The volume is so great that what's being published on COVID is equal to all the other research that is usually being put on infectious disease as a whole," says Uri Blackman, CEO of GIDEON Informatics, which has put out medical databases since 1992.
• It is impossible for a human being to keep up with all that science. Even if it only took an average of 15 minutes to read each paper, getting through all of them would require 287 days of nonstop reading, or more than 100 days longer than the outbreak itself.

How it works: That's where machine-learning tools come in.

• While current AI can't understand a scientific paper in the way that a trained human being can, it is capable of categorizing and ordering it in a way that reveals useful patterns. And it can do so much quicker than any human being can.
• The COVID-19 Primer takes in news articles and social media interactions that reference papers and their authors, which allows a user to quickly see which papers are getting the most attention from experts and average people alike.
• Papers can also be sorted through research categories like "patient and medical care" or "forecasts and modeling," and the algorithms behind the tool can also identify topics of interest that emerge from the entire corpus of research.

The catch: There are still major limits to AI's ability to actually comprehend written text of any kind, including scientific publications. That means the COVID-19 Primer and similar tools can't tell you the real value of any individual paper, just how it's being received in the news and on social media.

• In fact, says Amy Heineike of Prime AI, "the most shared papers are the most controversial ones." She notes that the second-most-shared paper since the start of the outbreak was a preprint publication from the end of January arguing the novel coronavirus appeared to contain genes from the HIV virus, with the implication that the virus was deliberately engineered.
• The paper — which like all preprints wasn't peer reviewed — was quickly torn apart by experts online, and was withdrawn by its authors just a few days after it was posted. But 20% of the tweets about the paper were posted in the days and weeks after its withdrawal, which was seen by many online as evidence of a conspiracy to suppress its conclusions.

The bottom line: Science is moving faster than ever on COVID-19, and AI can be a valuable tool to making sense of it all. But machines can't yet be a replacement for human judgment.

The possibility of employing geoengineering could help break the political deadlock on a global climate change deal, according to a new paper.

Why it matters: Deliberately trying to engineer the climate to offset warming is risky and as yet untested. But with the effects of climate change compounding and further international agreements stalled, there may be no choice but to try — or at least threaten to do so.

As climate change worsens and climate politics remain polarized, the door opens for solar geoengineering, which would involve injecting aerosols into the atmosphere to reduce the amount of sunlight reaching the Earth and directly slow warming.

• Solar geoengineering would be much cheaper than drastically cutting carbon emissions, but it carries with it as yet unknown risks of side effects. That's why many environmentalists remain opposed to even experimenting with geoengineering.

Yes, but: A new paper in the journal Humanities & Social Sciences Communications suggests that the possibility of geoengineering could reframe the climate debate.

• Using game theory (more or less as described in this scene from "A Beautiful Mind"), Gernot Wagner and Adrien Fabre make the case that countries most vulnerable to climate change and most willing to engage in deep emissions cuts might prefer trying geoengineering, even with all its risks, rather than accept insufficient climate action.
• At the same time, those countries less worried about climate change might be more willing to compromise on deeper emission cuts if they fear that the other side would choose geoengineering over a weaker climate deal.
• The upshot is that just the existence of the option of geoengineering can nudge the two opposing sides into agreeing on tougher climate action.

Yes, but: Game theory aside, the worse climate change gets, the more likely one or more countries might try geoengineering to save themselves. That's why we'd be better off ramping up experiments on geoengineering now, so we know better what it can do — and what it shouldn't do.

• "We need to know whether or not this should be part of the climate portfolio," says Kelly Wanser, the executive director of SilverLining, a nonprofit focused on averting near-term climate risk.

A startup is rolling out automated weed cutters at a moment when COVID-19 has made farm work more dangerous for human beings.

Why it matters: Robots in all fields have received a lift from the effects of the coronavirus pandemic, and farming is no different. One of the oldest occupations in the world could be the next to be automated.

There are two ways to remove weeds from the farms growing specialty crops like lettuce. One is to use herbicides — but such chemicals have increasingly come under fire for their alleged toxicity. The other is to employ workers to do the back-breaking labor of pulling out weeds by hand — but those workers have become harder to find due to shrinking labor pools and immigration restrictions.

• The startup FarmWise is offering a third option: robots. The company's Titan weed cutters autonomously trawl through fields, detecting and destroying weeds with retractable hoes.
• The robots use cameras and machine vision to create a 3D map of the farm plot and tell weeds from commercial plants.
• "One of the robots can replace the entire workflow for weeding, equivalent to a crew of 15 to 20 people," says Sébastien Boyer, FarmWise's CEO.

How it works: Like an increasing number of robot suppliers, FarmWise provides its Titan weed cutters — eight of which are currently operating in California's Salinas Valley — on a service model, charging farmers by the acre.

• Boyer sees FarmWise's robots eventually expanding to collect detailed data on farm plots, including everything from temperature to humidity to soil salinity. The result would be farming that is both automated and precise.
• With COVID-19 hitting migrant farm workers hard, robots could also be a safer option — albeit at the potential expense of those jobs.

What's next: For all their associations with nature, American farmers have been quick to adopt new technology, from the rotary combines of the 1970s to the self-steering, GPS-enabled tractors used today. With that technology has come the consolidation of farms and a drastic drop in the number of farmers.

The bottom line: With the annual shipment of mobile farm robots expected to grow by a hundredfold over the next decade, U.S. farms will become more efficient — and much emptier.

Things will never truly return to "normal" after the coronavirus. That's cause for eager anticipation, and also for dread, my Axios colleague Dave Lawler writes.

What to expect: The world after COVID-19 will be poorer, at least for a time. 

• It will be more unequal too, both among countries and within them — where many skimmed along without coffee meetings and business trips, while others performed newly dangerous jobs, or lost them.
• Developing nations were hit earliest and hardest, and they'll likely be the slowest to recover.

The other side: There's cause for optimism among the rubble. Practices and institutions that endured only through inertia before the pandemic may be halted, and then reversed.

• Systemic racism, inadequate health care and incompetent governance have all been laid bare. Unprecedented displays of international solidarity have filled streets from Minneapolis to Nairobi.
• Technologies and ideas that already existed are now being put to widespread use — to conduct business, education and health care remotely, for example.

What to watch: Innovation may surface in surprising ways. It was the death of thousands of horses in an 1815 famine, the Economist notes, that led Karl von Drais to invent the bicycle.

The bottom line: The pandemic won't only change the world, but also the way we look at it.

Go deeper

Why tech didn't save us from COVID-19 (David Rotman — MIT Tech Review)

• Drastic reduction in public funding for innovation left us vulnerable when a pandemic came calling.

Revenge of the suburbs (Ian Bogost — The Atlantic)

• No one seems to like the suburbs — except, of course, for all those people who live in them. Now with the pandemic making city life temporarily untenable, suburbanites are getting the last laugh.

The story of Juneteenth (Lynn Brown — JSTOR Daily)

• How the official end of American slavery — and the birth of what will hopefully be America's next national holiday — took place.

Inside China's DNA dragnet (Emile Dirks and James Leibold — Australian Strategic Policy Institute)

• A shocking report that details how the Chinese government is creating a police-run DNA database of tens of millions of its citizens.

The FDA this week approved a video game as a prescribable therapeutic for kids with ADHD.

Why it matters: The move marks the first time a video game can be legally marketed as a therapy for a health condition, and it shows the gradual progress of the wider field of digital therapeutics.

What's happening: The game, called EndeavorRx and developed by Akili Interactive Labs, rewards players with stars for navigating a fantasy landscape and finishing tasks.

• The game is meant to act as a delivery system for algorithms that can strengthen neural networks in the brain connected to ADHD, according to reporting by STAT.
• While Akili has made the game available for the past few months for free to eligible children, the FDA's decision means that physicians can now prescribe it, and insurers may cover it.

Context: The market for digital therapeutics — which range from games like EndeavorRx to apps focusing on addiction — has been growing in recent years. The pandemic gave the industry a push, as the lockdown led the FDA to relax regulations on a range of mental health and telemedicine apps.

• "We are seeing a 20% to 30% growth of people seeking help on anxiety and depression during the pandemic," says Adnan Asar, CEO of Lucid Lane, a digital health platform.

My thought bubble: I'm looking forward to the day when I can get Axios' health insurer to cover NBA 2K20.