The tricky math of lifting coronavirus lockdowns


James Temple at MIT Technology Review: “…A crucial point of the work—which Steinhardt and MIT’s Andrew Ilyas​ wrote up in a draft paper that hasn’t yet been published or peer-reviewed—is that communities need to get much better at tracking infections. “With the data we currently have, we actually just don’t know what the level of safe mobility is,” Steinhardt says. “We need much better mechanisms for tracking prevalence in order to do any of this safely.”

The analysis relies on other noisy and less-than-optimal measurements as well, including using hospitalization admissions and deaths to estimate disease prevalence before the lockdowns. They also had to make informed assumptions, which others might disagree with, about how much shelter-in-place rules have altered the spread of the disease. Much of the overall uncertainty is due to the spottiness of testing to date. If case counts are rising, but so is testing, it’s difficult to decipher whether infections are still increasing or a greater proportion of infected people are being evaluated.

This produces some confusing results in the study for any policymaker looking for clear direction. Notably, in Los Angeles, the estimated growth rate of the disease since the shelter-in-place order went into effect ranges from negative to positive. This suggests either that the city could start loosening restrictions or that it needs to tighten them further.

Ultimately, the researchers stress that communities need to build up disease surveillance measures to reduce this uncertainty, and strike an appropriate balance between reopening the economy and minimizing public health risks.

They propose several ways to do so, including conducting virological testing on a random sample of some 20,000 people per day in a given area; setting up wide-scale online surveys that ask people to report potential symptoms, similar to what Carnegie Mellon researchers are doing through efforts with both Facebook and Google; and potentially testing for the prevalence of viral material in wastewater, a technique that has “sounded the alarm” on polio outbreaks in the past.

A team of researchers affiliated with MIT, Harvard, and startup Biobot Analytics recently analyzed water samples from a Massachusetts treatment facility, and detected levels of the coronavirus that were “significantly higher” than expected on the basis of confirmed cases in the state, according to a non-peer-reviewed paper released earlier this month….(More)”.

Digital solutions to revolutionise community empowerment


Article by Alan Marcus: “…The best responses to Covid-19 have harmonised top-down policies and grassroots organisation. In the UK, more than 700,000 volunteers for the National Health Service are being organised through GoodSAM—an app that, like many gig economy platforms, allows individuals to switch on availability for delivering supplies to vulnerable people.

Perhaps the best example is Taiwan, where officials have kept the rate of infection to a fraction of even highly-rated Singapore. Coordinating public and private groups, the country has deployed a range of online services, including a system for mapping and allocating rationed face masks developed by Digital Minister Audrey Tang and members of an online hacktivist chatroom. …

Effective responses to the crisis show the value of inclusive government and hint at more resilient models for managing our communities. So far, governments, businesses and individuals have pooled resources to deliver country-wide responses. However, this model should be pushed further. Digital tools should be provided to communities to organise themselves, develop locally tailored solutions and get involved in the governance of their town or neighbourhood.

This model requires open communication between local people and the organisations responsible for administrating neighbourhoods—be they governments or businesses. … 

The platform provides significant opportunities for optimising crisis response and elevating quality of life. For example, a popular solution for market vendors forced to close by Covid-19 has been offering delivery services. As well as the businesses, this benefits local people, who can bypass overcrowded superstores or overcapacity online grocery deliveries. While grassroots movements are largely left to organise themselves, this is a missed opportunity for collaboration with local administrators.

By communicating with vendors, the administrator can not only establish an online platform to coordinate their services, but also connect them with local people to help deliver the service, such as van owners who can loan their vehicles. Moreover, the administrator can collect feedback on local infrastructure needed to improve services, such as communal cold lockers for receiving groceries when no-one is home.

By integrating this model into the day-to-day governance of our communities, we can unite community action with top-down resources, empowering local people to co-own the evolution of their neighbourhoods and helping administrators prioritise projects that maximise quality of life.

As Solnit wrote: “A disaster is a lot like a revolution when it comes to disruption and improvisation.” Pushed to their limits, countries are pioneering ways of coordinating local and national action. From this wave of innovation, we can empower communities to become more resilient in crises, more inclusive in their governance and more engaged in the determination of their future….(More)”.

German humanities scholars enlisted to end coronavirus lockdown


David Matthews at THE: “In contrast to other countries, philosophers, historians, theologians and jurists have played a major role advising the state as it seeks to loosen restrictions…

In the struggle against the new coronavirus, humanities academics have entered the fray – in Germany at least.

Arguably to a greater extent than has happened in the UK, France or the US, the country has enlisted the advice of philosophers, historians of science, theologians and jurists as it navigates the delicate ethical balancing act of reopening society while safeguarding the health of the public.

When the German federal government announced a slight loosening of restrictions on 15 April – allowing small shops to open and some children to return to school in May – it had been eagerly awaiting a report written by a 26-strong expert group containing only a minority of natural scientists and barely a handful of virologists and medical specialists.

Instead, this working group from the Leopoldina – Germany’s independent National Academy of Sciences dating back to 1652 – included historians of industrialisation and early Christianity, a specialist on the philosophy of law and several pedagogical experts.

This paucity of virologists earned the group a swipe from Markus Söder, minister-president of badly hit Bavaria, who has led calls in Germany for a tough lockdown (although earlier in the pandemic the Leopoldina did release a report written by more medically focused specialists).

But “the crisis is a complex one, it’s a systemic crisis” and so it needs to be dissected from every angle, argued Jürgen Renn, director of the Max Planck Institute for the History of Science, and one of those who wrote the crucial recommendations.

And Professor Renn – who earlier this year published a book on rethinking science in the Anthropocene – made the argument for green post-virus reconstruction. Urbanisation and deforestation have squashed mankind and wildlife together, making other animal-to-human disease transmissions ever more likely, he argued. “It’s not the only virus waiting out there,” he said.

Germany’s Ethics Council – which traces its roots back to the stem cell debates of the early 2000s and is composed of theologians, jurists, philosophers and other ethical thinkers – also contributed to a report at the end of March, warning that it was up to elected politicians, not scientists, to make the “painful decisions” weighing up the lockdown’s effect on health and its other side-effects….(More)“.

How data privacy leader Apple found itself in a data ethics catastrophe


Article by Daniel Wu and Mike Loukides: “…Apple learned a critical lesson from this experience. User buy-in cannot end with compliance with rules. It requires ethics, constantly asking how to protect, fight for, and empower users, regardless of what the law says. These strategies contribute to perceptions of trust.

Trust has to be earned, is easily lost, and is difficult to regain….

In our more global, diverse, and rapidly- changing world, ethics may be embodied by the “platinum rule”: Do unto others as they would want done to them. One established field of ethics—bioethics—offers four principles that are related to the platinum rule: nonmaleficence, justice, autonomy, and beneficence.

For organizations that want to be guided by ethics, regardless of what the law says, these principles as essential tools for a purpose-driven mission: protecting (nonmaleficence), fighting for (justice), and empowering users and employees (autonomy and beneficence).

An ethics leader protects users and workers in its operations by using governance best practices. 

Before creating the product, it understands both the qualitative and quantitative contexts of key stakeholders, especially those who will be most impacted, identifying their needs and fears. When creating the product, it uses data protection by design, working with cross-functional roles like legal and privacy engineers to embed ethical principles into the lifecycle of the product and formalize data-sharing agreements. Before launching, it audits the product thoroughly and conducts scenario planning to understand potential ethical mishaps, such as perceived or real gender bias or human rights violations in its supply chain. After launching, its terms of service and collection methods are highly readable and enables even disaffected users to resolve issues delightfully.

Ethics leaders also fight for users and workers, who can be forgotten. These leaders may champion enforceable consumer protections in the first place, before a crisis erupts. With social movements, leaders fight powerful actors preying on vulnerable communities or the public at large—and critically examines and ameliorates its own participation in systemic violence. As a result, instead of last-minute heroic efforts to change compromised operations, it’s been iterating all along.

Finally, ethics leaders empower their users and workers. With diverse communities and employees, they co-create new products that help improve basic needs and enable more, including the vulnerable, to increase their autonomy and their economic mobility. These entrepreneurial efforts validate new revenue streams and relationships while incubating next-generation workers who self-govern and push the company’s mission forward. Employees voice their values and diversify their relationships. Alison Taylor, the Executive Director of Ethical Systems, argues that internal processes should “improve [workers’] reasoning and creativity, instead of short-circuiting them.” Enabling this is a culture of psychological safety and training to engage kindly with divergent ideas.

These purpose-led strategies boost employee performance and retention, drive deep customer loyalty, and carve legacies.

To be clear, Apple may be implementing at least some of these strategies already—but perhaps not uniformly or transparently. For instance, Apple has implemented some provisions of the European Union’s General Data Protection Regulation for all US residents—not just EU and CA residents—including the ability to access and edit data. This expensive move, which goes beyond strict legal requirements, was implemented even without public pressure.

But ethics strategies have major limitations leaders must address

As demonstrated by the waves of ethical “principles” released by Fortune 500 companies and commissions, ethics programs can be murky, dominated by a white, male, and Western interpretation.

Furthermore, focusing purely on ethics gives companies an easy way to “free ride” off social goodwill, but ultimately stay unaccountable, given the lack of external oversight over ethics programs. When companies substitute unaccountable data ethics principles for thoughtful engagement with the enforceable data regulation principles, users will be harmed.

Long-term, without the ability to wave a $100 million fine with clear-cut requirements and lawyers trained to advocate for them internally, ethics leaders may face barriers to buy-in. Unlike their sales, marketing, or compliance counterparts, ethics programs do not directly add revenue or reduce costs. In recessions, these “soft” programs may be the first on the chopping block.

As a result of these factors, we will likely see a surge in ethics-washing: well-intentioned companies that talk ethics, but don’t walk it. More will view these efforts as PR-driven ethics stunts, which don’t deeply engage with actual ethical issues. If harmful business models do not change, ethics leaders will be fighting a losing battle….(More)”.

Tear down this wall: Microsoft embraces open data


The Economist: “Two decades ago Microsoft was a byword for a technological walled garden. One of its bosses called free open-source programs a “cancer”. That was then. On April 21st the world’s most valuable tech firm joined a fledgling movement to liberate the world’s data. Among other things, the company plans to launch 20 data-sharing groups by 2022 and give away some of its digital information, including data it has aggregated on covid-19.

Microsoft is not alone in its newfound fondness for sharing in the age of the coronavirus. “The world has faced pandemics before, but this time we have a new superpower: the ability to gather and share data for good,” Mark Zuckerberg, the boss of Facebook, a social-media conglomerate, wrote in the Washington Post on April 20th. Despite the EU’s strict privacy rules, some Eurocrats now argue that data-sharing could speed up efforts to fight the coronavirus. 

But the argument for sharing data is much older than the virus. The OECD, a club mostly of rich countries, reckons that if data were more widely exchanged, many countries could enjoy gains worth between 1% and 2.5% of GDP. The estimate is based on heroic assumptions (such as putting a number on business opportunities created for startups). But economists agree that readier access to data is broadly beneficial, because data are “non-rivalrous”: unlike oil, say, they can be used and re-used without being depleted, for instance to power various artificial-intelligence algorithms at once. 

Many governments have recognised the potential. Cities from Berlin to San Francisco have “open data” initiatives. Companies have been cagier, says Stefaan Verhulst, who heads the Governance Lab at New York University, which studies such things. Firms worry about losing intellectual property, imperilling users’ privacy and hitting technical obstacles. Standard data formats (eg, JPEG images) can be shared easily, but much that a Facebook collects with its software would be meaningless to a Microsoft, even after reformatting. Less than half of the 113 “data collaboratives” identified by the lab involve corporations. Those that do, including initiatives by BBVA, a Spanish bank, and GlaxoSmithKline, a British drugmaker, have been small or limited in scope. 

Microsoft’s campaign is the most consequential by far. Besides encouraging more non-commercial sharing, the firm is developing software, licences and (with the Governance Lab and others) governance frameworks that permit firms to trade data or provide access to them without losing control. Optimists believe that the giant’s move could be to data what IBM’s embrace in the late 1990s of the Linux operating system was to open-source software. Linux went on to become a serious challenger to Microsoft’s own Windows and today underpins Google’s Android mobile software and much of cloud-computing…(More)”.

The global pandemic has spawned new forms of activism – and they’re flourishing


Erica Chenoweth, Austin Choi-Fitzpatrick, Jeremy Pressman, Felipe G Santos and Jay Ulfelder at The Guardian: “Before the Covid-19 pandemic, the world was experiencing unprecedented levels of mass mobilization. The decade from 2010 to 2019 saw more mass movements demanding radical change around the world than in any period since World War II. Since the pandemic struck, however, street mobilization – mass demonstrations, rallies, protests, and sit-ins – has largely ground to an abrupt halt in places as diverse as India, Lebanon, Chile, Hong Kong, Iraq, Algeria, and the United States.

The near cessation of street protests does not mean that people power has dissipated. We have been collecting data on the various methods that people have used to express solidarity or adapted to press for change in the midst of this crisis. In just several weeks’ time, we’ve identified nearly 100 distinct methods of nonviolent action that include physical, virtual and hybrid actions – and we’re still counting. Far from condemning social movements to obsolescence, the pandemic – and governments’ responses to it – are spawning new tools, new strategies, and new motivation to push for change.

In terms of new tools, all across the world, people have turned to methods like car caravanscacerolazos (collectively banging pots and pans inside the home), and walkouts from workplaces with health and safety challenges to voice personal concerns, make political claims, and express social solidarity. Activists have developed alternative institutions such as coordinated mask-sewing, community mutual aid pods, and crowdsourced emergency funds. Communities have placed teddy bears in their front windows for children to find during scavenger hunts, authors have posted live-streamed readings, and musicians have performed from their balconies and rooftops. Technologists are experimenting with drones adapted to deliver supplies, disinfect common areas, check individual temperatures, and monitor high-risk areas. And, of course, many movements are moving their activities online, with digital ralliesteachins, and information-sharing.

Such activities have had important impacts. Perhaps the most immediate and life-saving efforts have been those where movements have begun to coordinate and distribute critical resources to people in need. Local mutual aid pods, like those in Massachusetts, have emerged to highlight urgent needs and provide for crowdsourced and volunteer rapid response. Pop-up food banks, reclaiming vacant housing, crowdsourced hardship funds, free online medical-consultation clinics, mass donations of surgical masks, gloves, gowns, goggles and sanitizer, and making masks at home are all methods that people have developed in the past several weeks. Most people have made these items by hand. Others have even used 3D printers to make urgently-needed medical supplies. These actions of movements and communities have already saved countless lives….(More)”.

The imperative of interpretable machines


Julia Stoyanovich, Jay J. Van Bavel & Tessa V. West at Nature: “As artificial intelligence becomes prevalent in society, a framework is needed to connect interpretability and trust in algorithm-assisted decisions, for a range of stakeholders.

We are in the midst of a global trend to regulate the use of algorithms, artificial intelligence (AI) and automated decision systems (ADS). As reported by the One Hundred Year Study on Artificial Intelligence: “AI technologies already pervade our lives. As they become a central force in society, the field is shifting from simply building systems that are intelligent to building intelligent systems that are human-aware and trustworthy.” Major cities, states and national governments are establishing task forces, passing laws and issuing guidelines about responsible development and use of technology, often starting with its use in government itself, where there is, at least in theory, less friction between organizational goals and societal values.

In the United States, New York City has made a public commitment to opening the black box of the government’s use of technology: in 2018, an ADS task force was convened, the first of such in the nation, and charged with providing recommendations to New York City’s government agencies for how to become transparent and accountable in their use of ADS. In a 2019 report, the task force recommended using ADS where they are beneficial, reduce potential harm and promote fairness, equity, accountability and transparency2. Can these principles become policy in the face of the apparent lack of trust in the government’s ability to manage AI in the interest of the public? We argue that overcoming this mistrust hinges on our ability to engage in substantive multi-stakeholder conversations around ADS, bringing with it the imperative of interpretability — allowing humans to understand and, if necessary, contest the computational process and its outcomes.

Remarkably little is known about how humans perceive and evaluate algorithms and their outputs, what makes a human trust or mistrust an algorithm3, and how we can empower humans to exercise agency — to adopt or challenge an algorithmic decision. Consider, for example, scoring and ranking — data-driven algorithms that prioritize entities such as individuals, schools, or products and services. These algorithms may be used to determine credit worthiness, and desirability for college admissions or employment. Scoring and ranking are as ubiquitous and powerful as they are opaque. Despite their importance, members of the public often know little about why one person is ranked higher than another by a résumé screening or a credit scoring tool, how the ranking process is designed and whether its results can be trusted.

As an interdisciplinary team of scientists in computer science and social psychology, we propose a framework that forms connections between interpretability and trust, and develops actionable explanations for a diversity of stakeholders, recognizing their unique perspectives and needs. We focus on three questions (Box 1) about making machines interpretable: (1) what are we explaining, (2) to whom are we explaining and for what purpose, and (3) how do we know that an explanation is effective? By asking — and charting the path towards answering — these questions, we can promote greater trust in algorithms, and improve fairness and efficiency of algorithm-assisted decision making…(More)”.

How Facebook and Google are helping the CDC forecast coronavirus


Karen Hao at MIT Technology Review: “When it comes to predicting the spread of an infectious disease, it’s crucial to understand what Ryan Tibshirani, an associate professor at Carnegie Mellon University, calls the “the pyramid of severity.” The bottom of the pyramid is asymptomatic carriers (those who have the infection but feel fine); the next level is symptomatic carriers (those who are feeling ill); then come hospitalizations, critical hospitalizations, and finally deaths.

Every level of the pyramid has a clear relationship to the next: “For example, sadly, it’s pretty predictable how many people will die once you know how many people are under critical care,” says Tibshirani, who is part of CMU’s Delphi research group, one of the best flu-forecasting teams in the US. The goal, therefore, is to have a clear measure of the lower levels of the pyramid, as the foundation for forecasting the higher ones.

But in the US, building such a model is a Herculean task. A lack of testing makes it impossible to assess the number of asymptomatic carriers. The results also don’t accurately reflect how many symptomatic carriers there are. Different counties have different testing requirements—some choosing only to test patients who require hospitalization. Test results also often take upwards of a week to return.

The remaining option is to measure symptomatic carriers through a large-scale, self-reported survey. But such an initiative won’t work unless it covers a big enough cross section of the entire population. Now the Delphi group, which has been working with the Centers for Disease Control and Prevention to help it coordinate the national pandemic response, has turned to the largest platforms in the US: Facebook and Google.

Facebook will help CMU Delphi research group gather data about Covid symptoms

In a new partnership with Delphi, both tech giants have agreed to help gather data from those who voluntarily choose to report whether they’re experiencing covid-like symptoms. Facebook will target a fraction of their US users with a CMU-run survey, while Google has thus far been using its Opinion Rewards app, which lets users respond to questions for app store credit. The hope is this new information will allow the lab to produce county-by-county projections that will help policymakers allocate resources more effectively.

Neither company will ever actually see the survey results; they’re merely pointing users to the questions administered and processed by the lab. The lab will also never share any of the raw data back to either company. Still, the agreements represent a major deviation from typical data-sharing practices, which could raise privacy concerns. “If this wasn’t a pandemic, I don’t know that companies would want to take the risk of being associated with or asking directly for such a personal piece of information as health,” Tibshirani says.

Without such cooperation, the researchers would’ve been hard pressed to find the data anywhere else. Several other apps allow users to self-report symptoms, including a popular one in the UK known as the Covid Symptom Tracker that has been downloaded over 1.5 million times. But none of them offer the same systematic and expansive coverage as a Facebook or Google-administered survey, says Tibshirani. He hopes the project will collect millions of responses each week….(More)”.

Tracking coronavirus: big data and the challenge to privacy


Nic Fildes and Javier Espinoza at the Financial Times: “When the World Health Organization launched a 2007 initiative to eliminate malaria on Zanzibar, it turned to an unusual source to track the spread of the disease between the island and mainland Africa: mobile phones sold by Tanzania’s telecoms groups including Vodafone, the UK mobile operator.

Working together with researchers at Southampton university, Vodafone began compiling sets of location data from mobile phones in the areas where cases of the disease had been recorded. 

Mapping how populations move between locations has proved invaluable in tracking and responding to epidemics. The Zanzibar project has been replicated by academics across the continent to monitor other deadly diseases, including Ebola in west Africa….

With much of Europe at a standstill as a result of the coronavirus pandemic, politicians want the telecoms operators to provide similar data from smartphones. Thierry Breton, the former chief executive of France Telecom who is now the European commissioner for the internal market, has called on operators to hand over aggregated location data to track how the virus is spreading and to identify spots where help is most needed.

Both politicians and the industry insist that the data sets will be “anonymised”, meaning that customers’ individual identities will be scrubbed out. Mr Breton told the Financial Times: “In no way are we going to track individuals. That’s absolutely not the case. We are talking about fully anonymised, aggregated data to anticipate the development of the pandemic.”

But the use of such data to track the virus has triggered fears of growing surveillance, including questions about how the data might be used once the crisis is over and whether such data sets are ever truly anonymous….(More)”.

Coronavirus: country comparisons are pointless unless we account for these biases in testing


Norman Fenton, Magda Osman, Martin Neil, and Scott McLachlan at The Conversation: “Suppose we wanted to estimate how many car owners there are in the UK and how many of those own a Ford Fiesta, but we only have data on those people who visited Ford car showrooms in the last year. If 10% of the showroom visitors owned a Fiesta, then, because of the bias in the sample, this would certainly overestimate the proportion of Ford Fiesta owners in the country.

Estimating death rates for people with COVID-19 is currently undertaken largely along the same lines. In the UK, for example, almost all testing of COVID-19 is performed on people already hospitalised with COVID-19 symptoms. At the time of writing, there are 29,474 confirmed COVID-19 cases (analogous to car owners visiting a showroom) of whom 2,352 have died (Ford Fiesta owners who visited a showroom). But it misses out all the people with mild or no symptoms.

Concluding that the death rate from COVID-19 is on average 8% (2,352 out of 29,474) ignores the many people with COVID-19 who are not hospitalised and have not died (analogous to car owners who did not visit a Ford showroom and who do not own a Ford Fiesta). It is therefore equivalent to making the mistake of concluding that 10% of all car owners own a Fiesta.

There are many prominent examples of this sort of conclusion. The Oxford COVID-19 Evidence Service have undertaken a thorough statistical analysis. They acknowledge potential selection bias, and add confidence intervals showing how big the error may be for the (potentially highly misleading) proportion of deaths among confirmed COVID-19 patients.

They note various factors that can result in wide national differences – for example the UK’s 8% (mean) “death rate” is very high compared to Germany’s 0.74%. These factors include different demographics, for example the number of elderly in a population, as well as how deaths are reported. For example, in some countries everybody who dies after having been diagnosed with COVID-19 is recorded as a COVID-19 death, even if the disease was not the actual cause, while other people may die from the virus without actually having been diagnosed with COVID-19.

However, the models fail to incorporate explicit causal explanations in their modelling that might enable us to make more meaningful inferences from the available data, including data on virus testing.

What a causal model would look like. Author provided

We have developed an initial prototype “causal model” whose structure is shown in the figure above. The links between the named variables in a model like this show how they are dependent on each other. These links, along with other unknown variables, are captured as probabilities. As data are entered for specific, known variables, all of the unknown variable probabilities are updated using a method called Bayesian inference. The model shows that the COVID-19 death rate is as much a function of sampling methods, testing and reporting, as it is determined by the underlying rate of infection in a vulnerable population….(More)”