Thanks in part to pandemic-driven disruptions of conventional meat processing, sales and interest in plant-based alternatives are taking off, changing the future of food.

Why it matters: Meat-processing plants have proven especially vulnerable to coronavirus outbreaks, and meat consumption adds to climate change. Better-tasting alternatives could shrink that environmental footprint while solidifying the supply chain for protein.

Driving the news: The UN Food and Agriculture Organization reported this week that global per capita meat consumption is projected to fall to the lowest level in nine years.

• In the U.S., per capita meat consumption isn't expected to return to pre-pandemic levels until at least 2025.

By the numbers: While demand for conventional meat has fallen, new plant-based alternatives are on the rise.

• Grocery store sales of alternative-meat products rose 264% in the nine weeks ending May 2 — faster than they were growing in the weeks before the pandemic, according to data from Nielsen,
• Impossible Foods, a leading maker of alternative meat, reports its grocery store footprint has increased 18-fold since March, and it expects to see a 50-fold rise by the end of 2020.
• Last month JBS, the world's biggest meat seller, launched its own brand of plant-based burgers and chorizo alternatives, a sign that conventional producers want in on plant-based business.

Some environmentalists criticize the most popular alternative-meat products for being overly processed. There's some truth to that charge, but that processed quality has made the alternative sector more resilient to the coronavirus.

• As the Wall Street Journal reported this week, more than 17,300 meat and poultry workers were infected with COVID-19 in April and May, in part because meat processors employ 3.2 workers per 1,000 sq. ft. of manufacturing space — three times the national average for manufacturers.
• Alternative-protein production tends to be significantly more automated than conventional meat processing, and fewer workers means less vulnerability to outbreaks.
• "All our focus is about enhancing the efficiency of what we do," says Thomas Jonas, the CEO of the Chicago-based alternative-protein company Nature's Fynd. "Doing more with less — that should be the motto of our species for the next few decades."

Details: This spring Nature's Fynd began production at a 35,000 sq. ft. manufacturing facility on the site of Chicago's Union Stockyards, a historic meatpacking center and the location for Upton Sinclair's notorious exposé "The Jungle."

• The company produces alternative protein using microbes discovered during research into Yellowstone's hot springs. Nature's Fynd says that at full capacity, it will be able to make as many hamburger substitutes in a year as 16,000 acres of grazing cows, with just 1% of the equivalent greenhouse gases.

The catch: Despite the recent growth, the alternative-meat sector is still a fraction of the size of the conventional meat industry.

• Conventional meat producers are also investing heavily in automation technology that should make their facilities more resilient to pandemics and labor shortages.

The bottom line: COVID-19 could mark a lasting disruption to the most American of foods: the burger.

What's happening: The 3% decline from 2019 is in part a result of the pandemic-caused economic slowdown, as people around the world cut more expensive meat out of their diet. Beef prices in the U.S. rose 10% in May, largely due to widespread outbreaks of COVID-19 in meat-processing plants.

• The situation is far worse in the poorest developing nations, where the coronavirus and resulting lockdowns have devastated food supplies.
• In a new report, Oxfam International estimates that up to 12,000 people per day could die from hunger linked to the pandemic by the end of the year.

Researchers are calling for open and free access to U.S. court records and building an AI tool to analyze them.

Why it matters: Court records are publicly available but expensive to access and difficult to navigate. Freeing up that data — and using machine learning tools to make sense of it — would help make the justice system more just.

While records for Congress and executive agencies are free on the internet, federal courts charge $0.10 per printed page to view any record online.

• That makes it difficult and costly for researchers, journalists and ordinary citizens to tap the raw data needed to understand the inner workings of the U.S. justice system.

What's new: In one example of the kind of analysis that could be possible with open access, researchers from Northwestern University used an algorithm to scan court records and determine how often judges granted waivers for the $400 fee required to file a federal lawsuit.

• While there is no uniform standard for granting waivers, the researchers found unexpectedly huge variations. In one district, the approval rate varied from less than 20% for some judges to more than 80% for others.
• With open access "we can get a fuller picture of what the systematic trends are and make it all easily accessible," says Adam Pah of Northwestern's Kellogg School of Management and Organizations and one of the co-authors of the study.

What's next: The Northwestern researchers are working on an AI-powered platform called SCALES-OKN that would make federal courtroom data accessible to the public and easily analyzable, linking data in the courts to information outside them.

• Such a platform could be a potent tool for uncovering hidden bias over money or race in the justice system, says Pah.

The big picture: AI is already being used in the criminal justice system for policing and sentencing, but experts say it too often perpetuates a biased system. Unleashing AI on open court records could provide a welcome opportunity to use technology to further justice, not curtail it.

A coalition of AI groups has formed to produce a comprehensive data source on the coronavirus pandemic for policymakers and health care leaders.

Why it matters: A torrent of data about COVID-19 is being produced, but unless it can be organized in an accessible format, it will do little good. The new initiative aims to use machine learning and human expertise to produce meaningful insights for an unprecedented situation.

Driving the news: Members of the newly formed Collective and Augmented Intelligence Against COVID-19 (CAIAC) announced on Thursday include the Future Society, a nonprofit think tank from the Harvard Kennedy School of Government, as well as the Stanford Institute for Human-Centered Artificial Intelligence and representatives from UN agencies.

• Within six to eight weeks, CAIAC's platform will produce a decision-making tool that will initially focus on digital contact tracing of coronavirus infections, ferreting out misinformation about the pandemic and identifying second- and third-order effects of COVID-19 that go beyond illness and death.

What they're saying: "With COVID-19 we realized there are tons of data available, but there was little global coordination on how to share it," says Cyrus Hodes, chair of the AI Initiative at the Future Society and a member of the CAIAC steering committee. "That's why we created this coalition — to put together a sense-making platform for policymakers to use."

• Hodes says that CAIAC will be targeting U.N. agencies as some of the first users for the platform.

The bottom line: Humans aren't exactly doing a great job beating COVID-19, so we need all the machine help we can get.

The spillover of pathogens from animals to humans — driven mainly by human behaviors like urbanization and the demand to eat meat — is increasing and will continue wreaking havoc unless global action is taken, Axios' Eileen Drage O'Reilly reports.

What can be done: Several scientific experts tell Axios global surveillance is one of the big first steps to prevention.

1. Track wildlife and livestock to catch pathogens before they spread to humans.

• One suggestion is reinstating PREDICT, a federally funded program designed to hunt down new pandemic pathogens globally that was shut down in 2019 just before the COVID-19 pandemic, despite its now-prescient warnings.

2. Fund and facilitate new technologies that can provide efficient, less costly and consistent testing, even in remote areas.

• "There is a scarcity of genetic and genomic research technologies within the conservation world and within ecology, at a lot of these really biodiverse regions. And we think that this decentralized model can really bring that technology to those areas," says Caroline Moore, a fellow with San Diego Zoo Global in disease investigations and part of the focus group.

3. Develop global databases to first track what's happening in wildlife and then eventually with humans.

Go deeper.

The pandemic and the dawn of an 'Asian century' (Ishaan Tharoor — Washington Post)

• America's bungled response to COVID-19 is seriously damaging the country's international image, even as Asian nations capitalize on the pandemic.

Slate Star Codex and Silicon Valley's war against the media (Gideon Lewis-Kraus — The New Yorker)

• A deep dive into the controversy over a rationalist blog and the growing hostility between the tech community and the reporters who cover it.

The damage we're not attending to (David Krakauer and Geoffrey West — Nautilus)

• As we focus on the virus, we risk missing the damage the pandemic is doing to the connected complex systems that determine how we live.

Why general artificial intelligence will not be realized (Ragnar Fjelland — Nature)

• Well, the good news is that if the author is right, at least we never have to worry about this.

A team of scientists has developed a technique that automatically makes written sentences more polite.

Why it matters: As the authors themselves note in the paper, it is "imperative to use the appropriate level of politeness for smooth communication in conversations." And what better to determine the appropriate level of politeness than an unfeeling machine-learning algorithm?

What's new: In a paper presented this week at the annual meeting of the Association for Computational Linguistics, researchers from Carnegie Mellon University analyzed a dataset of 1.39 million sentences, each of which was labeled with a politeness score.

• Using what is called a "tag and generate" approach, those sentences labeled as impolite were tagged, and then new text was generated to make the phrase nicer.
• So a phrase like "send me the data" would be automatically restructured to read "could you please send me the data?"
• Sometimes the system veered into outright editorializing, changing a sentence like "their chips are ok" to "their chips are great," which I must say may be more polite but could inadvertently end up sending someone to a mediocre Mexican restaurant.

Of note: The researchers used the "Enron Corpus" as a dataset — hundreds of thousands of emails exchanged by Enron employees and preserved by the federal government during its investigation of the now-defunct energy firm.

• A copy of the Enron Corpus was purchased by a computer scientist and released to researchers, who have since used it for numerous machine-learning studies.
• But really, why not train an AI for social niceties on the internal messages of a company that carried out one of the greatest frauds in American history?

The bottom line: This paper is really, really fascinating, and I am absolutely not just being polite.