COVID-19 appears to be accelerating the adoption of workplace automation — and the trend is likely to stick around after the pandemic.

Why it matters: Adopting robots and AI could keep businesses going during social distancing and reduce the health risk to human workers. But with unemployment already at Great Depression levels, many of the jobs lost to automation might never be regained.

What's happening: Brain Corp, a San Diego-based company that develops software for use in autonomous cleaning robots, reports its customers are employing robots about 13% more than they were in the months before the pandemic.

• The autonomous cleaners can do basic cleaning tasks "so that workers can use their time to do essential sanitation," says Phil Duffy, vice president of innovation at Brain Corp. "Robots are something a lot of our customers are looking at now, and it's making a big difference."
• Simbe Robotics produces an autonomous shelf-scanning robot called Tally that can audit inventory at grocery stores through computer vision and machine learning. That's particularly useful for food markets as they struggle to keep products on the shelf during the disruption of the pandemic, says Brad Bogolea, Simbe's CEO.
• Fetch Robotics employs a cloud platform that allows for the rapid deployment of robots in warehouses and similar facilities. With fewer human workers on the job because of social distancing, essential businesses "need robots to help make up the difference," says Melonee Wise, Fetch's CEO.

The big picture: Past experience suggests the advance of automation happens in sudden surges — and economic downturns are often a trigger.

• A 2018 study looked at three recessions over the past 30 years and found 88% of the jobs lost were in routine, "automatable" occupations.
• A 2016 paper examined almost 100 million job posts online before and after the 2008 recession and concluded companies in hard-hit areas were replacing employees performing routine tasks with a mix of technology and more skilled workers.

What to watch: Robots will be particularly attractive to front-line businesses that have to stay open during the pandemic.

"When you have the effect of social distancing and infection, it adds to the rationale for the substitution of machines for humans at point-of-sale positions like a cashier for food service, or in hospitals."

Yes, but: Companies in the robotics business say their products are meant to augment human workers, not replace them. But with tens of millions of Americans unemployed, it's impossible not to fear that a surge in automation could make a post-pandemic job recovery even more difficult.

• Low-income jobs will be particularly vulnerable to automation. As a result, "automation and digital technologies are exacerbating social cleavages and could be a source of unrest for years to come," wrote Carl Benedikt Frey of Oxford's Future of Work program for the Financial Times.

The other side: Some experts believe the immediate threat to jobs from automation during the pandemic is overstated.

• A January paper noted that in the past, automation rarely replaced entire occupational classes. Instead, robots tended to automate parts of a job, which can lead to a loss of income but not necessarily unemployment.
• Humans are still far more flexible, mentally and physically, than industrial robots, and their skills are still in demand. Amazon, which hasn't been shy about employing automation where it can, is looking to hire 100,000 human workers in its fulfillment centers and as delivery drivers.

The bottom line: The robots were already coming for jobs, and the pandemic will give employers additional incentive to automate where they can. But for now, the far bigger threat to jobs is the brute fact of an economic depression.

A recent report by Deloitte posits four potential scenarios for how the COVID-19 pandemic could affect the economy and society.

Why it matters: It's almost impossible to predict how an event as unprecedented as the pandemic will play out. But scenario planning allows business leaders to identify the most important questions a crisis poses, and prepare for a number of possible outcomes, rather than being locked into one future.

• To make some sense of how the next three to five years will look, Deloitte convened a number of top scenario thinkers in an exercise led by Andrew Blau, a managing director at the firm. I spoke to Blau about the four scenarios that resulted.

The Passing Storm: After a slow start, the pandemic is contained through an increasingly effective health care system and political response.

• Blau: "It feels unlikely at this point, but the surprise scenario is the one people are always least prepared for. It would be a storm that leaves us marked, but it wouldn't change everything."

Good Company: Governments struggle to control the pandemic. Large companies step up in their stead, accelerating the trend toward a more empathetic stakeholder capitalism.

• Blau: "Companies need to imagine a world where customers are going to make very different demands on them, one where their relationship to governments will be fundamentally reshaped by this experience."

Sunrise in the East: Western countries like the U.S. struggle to manage the pandemic compared to China and other East Asian nations. As a result, Beijing seizes geopolitical primacy.

• Blau: "This draws on a trend that was already in place before the pandemic: the continued rise of China. This unprecedented situation could take that trend and accelerate it.

Lone Wolves: The pandemic lasts longer than anyone expects, and in response, governments turn isolationist and tech-enabled surveillance becomes more common.

• Blau: "This is the scenario that no one wants, but sometimes, that's the outcome that happens. It's a world in which it seems we can't eradicate the disease, and social distancing becomes a way of life.

The bottom line: "My one prediction is that the future will surprise us," Blau told me. But the scenarios outlined here at least give us a platform on which to try to prepare.

A comprehensive new assessment of insect diversity finds that while the overall population of land-dwelling insects has fallen by more than a quarter over the past 30 years, some species are increasing in numbers.

Why it matters: A raft of studies in recent years have raised alarms about an "insect apocalypse." The new assessment offers some room for hope, while making it clear that insects and arachnid populations are still under tremendous pressure.

The new meta-analysis, published in this week's Science, examined 166 long-term surveys of land-dwelling and freshwater insect populations across the globe.

• Insects that live on land are struggling, declining by an average of 9% a decade, likely due in part to the spread of human populations. That's still a smaller decline than many earlier studies had found.
• Freshwater insect populations appear to be increasing by an average of 11% per decade, which may be due to successful efforts to clean up rivers and lakes.
• Insect declines are worse in North America, and especially the Midwest, but appear to be leveling off.

"Insect populations are like logs of wood that are pushed under water. They want to come up, while we keep pushing them farther down. But we can reduce the pressure so they can rise again."

The bottom line: Insects are an invaluable part of the Earth's ecosystem and food web. Their future is tied into ours — and vice versa.

Whether you work in a factory, a retail store, a restaurant or an office, you're going to have to get used to wearing a mask at work for the foreseeable future, my Axios colleague Joann Muller writes.

Context: I visited a former GM transmission factory Thursday that is now a hub of mask-making activity.

To be allowed inside, I had to practice all of the new health safety protocols that GM is instituting at its factories and that are likely to be similar for any workplace.

• I sanitized my hands and then put on a mask.
• I had my temperature taken and answered a health questionnaire.
• I did not sign in. Instead the security guard signed me in from behind a cordoned-off visitors desk.
• Inside the clean room where GM is making the masks, I donned a gown and a hairnet for added precautions.

Gerald Johnson, GM's executive vice president of global manufacturing, says people just have to get used to it.

• "It took us decades to learn how to wear seat belts. Today nobody questions it."

Go deeper

'Immunity passports' could create a new category of privilege (Emily Mullin — OneZero)

• If COVID-19 antibody tests indicate immunity to the disease — which isn't certain — they could be misused to discriminate in the workplace against the untested.

All the things COVID-19 will change forever, according to 30 top experts (Mark Sullivan — Fast Company)

• From virtual education to virtual working, get your coronavirus futurist takes wholesale with this piece.

Why we need worst-case thinking to prevent pandemics (Toby Ord — Guardian)

• An Oxford existential risk scholar, whom I interviewed recently, on why COVID-19 reminds us of the need to prepare for low-probability but high-consequence risks.

The pandemic can be a catalyst for decolonization in Africa (David Mwambari — Al Jazeera)

• A fascinating piece that looks at how Africa, so long bedeviled by infectious disease, could use the pandemic as an opportunity to chart an independent path.

An Italian-based artificial intelligence company is regularly analyzing social media posts about the coronavirus for their emotional content.

Why it matters: Classifying tens of thousands of posts by their emotional tone provides a snapshot of how people feel about the pandemic. Spoiler alert: not great!

How it works: Expert System specializes in semantics and natural language reading, a branch of AI involving computer systems that attempt to make sense of written language.

• In doing so, a computer can rapidly analyze vast amounts of the written word — like, for example, a day's worth of social media posts about the COVID-19 pandemic.

For the past few weeks, Expert System has been collecting English language social media posts each day that feature frequently used hashtags like #coronalockdown and #covid19. Its AI can extract the emotional content of those posts, which is then analyzed and interpreted by Sociometrica.

• On April 24, "fear" had become the single most widespread emotion, displacing "sadness."

But, but, but: Such negative feelings have been declining over the past 10 days, from 62.4% to 45.5%. At the same time, neutral and positive feelings are on the rise, with particular growth around posts showing "hope."

• Also increasing in intensity is "health fanaticism," which Expert System defines as "a feeling of fear and anxiety around certain aspects of health and an emphasis on defending the health of one’s own body."

The bottom line: A natural language AI can tell you what you probably already know: the pandemic is terrible, but if you squint hard enough, there's light at the end of the tunnel.