Odd Numbers: Algorithms alone can’t meaningfully hold other algorithms accountable

Curated on Posted on by Stefaan Verhulst

Frank Pasquale at Real Life Magazine: “Algorithms increasingly govern our social world, transforming data into scores or rankings that decide who gets credit, jobs, dates, policing, and much more. The field of “algorithmic accountability” has arisen to highlight the problems with such methods of classifying people, and it has great promise: Cutting-edge work in critical algorithm studies applies social theory to current events; law and policy experts seem to publish new articles daily on how artificial intelligence shapes our lives, and a growing community of researchers has developed a field known as “Fairness, Accuracy, and Transparency in Machine Learning.”

The social scientists, attorneys, and computer scientists promoting algorithmic accountability aspire to advance knowledge and promote justice. But what should such “accountability” more specifically consist of? Who will define it? At a two-day, interdisciplinary roundtable on AI ethics I recently attended, such questions featured prominently, and humanists, policy experts, and lawyers engaged in a free-wheeling discussion about topics ranging from robot arms races to computationally planned economies. But at the end of the event, an emissary from a group funded by Elon Musk and Peter Thiel among others pronounced our work useless. “You have no common methodology,” he informed us (apparently unaware that that’s the point of an interdisciplinary meeting). “We have a great deal of money to fund real research on AI ethics and policy”— which he thought of as dry, economistic modeling of competition and cooperation via technology — “but this is not the right group.” He then gratuitously lashed out at academics in attendance as “rent seekers,” largely because we had the temerity to advance distinctive disciplinary perspectives rather than fall in line with his research agenda.

Most corporate contacts and philanthrocapitalists are more polite, but their sense of what is realistic and what is utopian, what is worth studying and what is mere ideology, is strongly shaping algorithmic accountability research in both social science and computer science. This influence in the realm of ideas has powerful effects beyond it. Energy that could be put into better public transit systems is instead diverted to perfect the coding of self-driving cars. Anti-surveillance activism transmogrifies into proposals to improve facial recognition systems to better recognize all faces. To help payday-loan seekers, developers might design data-segmentation protocols to show them what personal information they should reveal to get a lower interest rate. But the idea that such self-monitoring and data curation can be a trap, disciplining the user in ever finer-grained ways, remains less explored. Trying to make these games fairer, the research elides the possibility of rejecting them altogether….(More)”.

World War Web

Curated on Posted on by Stefaan Verhulst

Special issue of Foreign Affairs: “The last few decades have witnessed the growth of an American-sponsored Internet open to all. But that was then; conditions have changed.

History is filled with supposed lost utopias, and there is no greater cliché than to see one’s own era as a lamentable decline from a previous golden age. Sometimes, however, clichés are right. And as we explored the Internet’s future for this issue’s lead package, it became clear this was one of those times. Contemplating where we have come from digitally and where we are heading, it’s hard not to feel increasingly wistful and nostalgic.

The last few decades have witnessed the growth of an American-sponsored Internet open to all, and that has helped tie the world together, bringing wide-ranging benefits to billions. But that was then; conditions have changed.

Other great powers are contesting U.S. digital leadership, pushing their own national priorities. Security threats appear and evolve constantly. Platforms that were supposed to expand and enrich the marketplace of ideas have been hijacked by trolls and bots and flooded with disinformation. And real power is increasingly concentrated in the hands of a few private tech giants, whose self-interested choices have dramatic consequences for the entire world around them.

Whatever emerges from this melee, it will be different from, and in many ways worse than, what we have now.

Adam Segal paints the big picture well. “The Internet has long been an American project,” he writes. “Yet today, the United States has ceded leadership in cyberspace to China.” What will happen if Beijing continues its online ascent? “The Internet will be less global and less open. A major part of it will run Chinese applications over Chinese-made hardware. And Beijing will reap the economic, diplomatic, national security, and intelligence benefits that once flowed to Washington.”

Nandan Nilekani, a co-founder of Infosys, outlines India’s unique approach to these issues, which is based on treating “digital infrastructure as a public good and data as something that citizens deserve access to.” Helen Dixon, Ireland’s data protection commissioner, presents a European perspective, arguing that giving individuals control over their own data—as the General Data Protection Regulation, the EU’s historic new regulatory effort, aims to do—is essential to restoring the Internet’s promise. And Karen Kornbluh, a veteran U.S. policymaker, describes how the United States dropped the digital ball and what it could do to pick it up again.

Finally, Michèle Flournoy and Michael Sulmeyer explain the new realities of cyberwarfare, and Viktor Mayer-Schönberger and Thomas Ramge consider the problems caused by Big Tech’s hoarding of data and what can be done to address it.

A generation from now, people across the globe will no doubt revel in the benefits the Internet has brought. But the more thoughtful among them will also lament the eclipse of the founders’ idealistic vision and dream of a world connected the way it could—and should— have been….(More)”.

As democracy goes digital, those offline are being pushed out of politics

Curated on Posted on by Stefaan Verhulst

Renata Avila at the Web Foundation: “Free and fair elections require an informed, active body of citizens debating the electoral issues of the day and scrutinising the positions of candidates. Participation at each and every stage of an electoral campaign — not just on the day of the vote — is necessary for a healthy democracy.

Those online have access to an increasingly sophisticated set of tools to do just this: to learn about candidates, to participate in political discussions, to shape debate and raise issues that matter to them. Or even, run for office themselves.

What does this mean for those citizens who don’t have access to the internet? Do online debates capture their needs, concerns and interests? Are the priorities of those not connected represented on the political stage?

The Mexican election: a story of digital inequality

María de Jesús “Marichuy” Patricio Martinez was selected as an independent candidate in Mexico’s recent July 1 elections general election — the first indigenous woman to run for president. But digital barriers doomed her candidacy.

Independent presidential candidates in Mexico are required to collect 866,000 signatures using a mandatory mobile app that only runs on relatively new smartphones. This means that to collect the required endorsements, a candidate and their supporters all need a modern smartphone — which typically costs around three times the minimum monthly salary — plus electricity and mobile data. These are resources many people in indigenous communities simply don’t have. While the electoral authorities exempted some municipalities from this process, it did not cover the mostly poor and indigenous areas that Marichuy wanted to represent. She was unable to gather the signatures needed….(More)”.

Searching for the Smart City’s Democratic Future

Curated on Posted on by Stefaan Verhulst

Article by Bianca Wylie at the Center for International Governance Innovation: “There is a striking blue building on Toronto’s eastern waterfront. Wrapped top to bottom in bright, beautiful artwork by Montreal illustrator Cecile Gariepy, the building — a former fish-processing plant — stands out alongside the neighbouring parking lots and a congested highway. It’s been given a second life as an office for Sidewalk Labs — a sister company to Google that is proposing a smart city development in Toronto. Perhaps ironically, the office is like the smart city itself: something old repackaged to be light, fresh and novel.

“Our mission is really to use technology to redefine urban life in the twenty-first century.”

Dan Doctoroff, CEO of Sidewalk Labs, shared this mission in an interview with Freakonomics Radio. The phrase is a variant of the marketing language used by the smart city industry at large. Put more simply, the term “smart city” is usually used to describe the use of technology and data in cities.

No matter the words chosen to describe it, the smart city model has a flaw at its core: corporations are seeking to exert influence on urban spaces and democratic governance. And because most governments don’t have the policy in place to regulate smart city development — in particular, projects driven by the fast-paced technology sector — this presents a growing global governance concern.

This is where the story usually descends into warnings of smart city dystopia or failure. Loads of recent articles have detailed the science fiction-style city-of-the-future and speculated about the perils of mass data collection, and for good reason — these are important concepts that warrant discussion. It’s time, however, to push past dystopian narratives and explore solutions for the challenges that smart cities present in Toronto and globally…(More)”.

Data-Driven Law: Data Analytics and the New Legal Services

Curated on Posted on by Stefaan Verhulst

Book by Edward J. Walters: “For increasingly data-savvy clients, lawyers can no longer give “it depends” answers rooted in anecdata. Clients insist that their lawyers justify their reasoning, and with more than a limited set of war stories. The considered judgment of an experienced lawyer is unquestionably valuable. However, on balance, clients would rather have the considered judgment of an experienced lawyer informed by the most relevant information required to answer their questions.

Data-Driven Law: Data Analytics and the New Legal Services helps legal professionals meet the challenges posed by a data-driven approach to delivering legal services. Its chapters are written by leading experts who cover such topics as:

  • Mining legal data
  • Computational law
  • Uncovering bias through the use of Big Data
  • Quantifying the quality of legal services
  • Data mining and decision-making
  • Contract analytics and contract standards

In addition to providing clients with data-based insight, legal firms can track a matter with data from beginning to end, from the marketing spend through to the type of matter, hours spent, billed, and collected, including metrics on profitability and success. Firms can organize and collect documents after a matter and even automate them for reuse. Data on marketing related to a matter can be an amazing source of insight about which practice areas are most profitable….(More)”.

Data Publics: Urban Protest, Analytics and the Courts

Curated on Posted on by Stefaan Verhulst

Article by Anthony McCosker and Timothy Graham in MC Journal: “There are many examples globally of the use of social media to engage publics in battles over urban development or similar issues (e.g. Fredericks and Foth). Some have asked how social media might be better used by neighborhood organisations to mobilise protest and save historic buildings, cultural landmarks or urban sites (Johnson and Halegoua). And we can only note here the wealth of research literature on social movements, protest and social media. To emphasise Gerbaudo’s point, drawing on Mattoni, we “need to account for how exactly the use of these media reshapes the ‘repertoire of communication’ of contemporary movements and affects the experience of participants” (2). For us, this also means better understanding the role that social data plays in both aiding and reshaping urban protest or arming third sector groups with evidence useful in social institutions such as the courts.

New modes of digital engagement enable forms of distributed digital citizenship, which Meikle sees as the creative political relationships that form through exercising rights and responsibilities. Associated with these practices is the transition from sanctioned, simple discursive forms of social protest in petitions, to new indicators of social engagement in more nuanced social media data and the more interactive forms of online petition platforms like change.org or GetUp (Halpin et al.). These technical forms code publics in specific ways that have implications for contemporary protest action. That is, they provide the operational systems and instructions that shape social actions and relationships for protest purposes (McCosker and Milne).

All protest and social movements are underwritten by explicit or implicit concepts of participatory publics as these are shaped, enhanced, or threatened by communication technologies. But participatory protest publics are uneven, and as Kelty asks: “What about all the people who are neither protesters nor Twitter users? In the broadest possible sense this ‘General Public’ cannot be said to exist as an actual entity, but only as a kind of virtual entity” (27). Kelty is pointing to the porous boundary between a general public and an organised public, or formal enterprise, as a reminder that we cannot take for granted representations of a public, or the public as a given, in relation to Like or follower data for instance.

If carefully gauged, the concept of data publics can be useful. To start with, the notions of publics and publicness are notoriously slippery. Baym and boyd explore the differences between these two terms, and the way social media reconfigures what “public” is. Does a Comment or a Like on a Facebook Page connect an individual sufficiently to an issues-public? As far back as the 1930s, John Dewey was seeking a pragmatic approach to similar questions regarding human association and the pluralistic space of “the public”. For Dewey, “the machine age has so enormously expanded, multiplied, intensified and complicated the scope of the indirect consequences [of human association] that the resultant public cannot identify itself” (157). To what extent, then, can we use data to constitute a public in relation to social protest in the age of data analytics?

There are numerous well formulated approaches to studying publics in relation to social media and social networks. Social network analysis (SNA) determines publics, or communities, through links, ties and clustering, by measuring and mapping those connections and to an extent assuming that they constitute some form of sociality. Networked publics (Ito, 6) are understood as an outcome of social media platforms and practices in the use of new digital media authoring and distribution tools or platforms and the particular actions, relationships or modes of communication they afford, to use James Gibson’s sense of that term. “Publics can be reactors, (re)makers and (re)distributors, engaging in shared culture and knowledge through discourse and social exchange as well as through acts of media reception” (Ito 6). Hashtags, for example, facilitate connectivity and visibility and aid in the formation and “coordination of ad hoc issue publics” (Bruns and Burgess 3). Gray et al., following Ruppert, argue that “data publics are constituted by dynamic, heterogeneous arrangements of actors mobilised around data infrastructures, sometimes figuring as part of them, sometimes emerging as their effect”. The individuals of data publics are neither subjugated by the logics and metrics of digital platforms and data structures, nor simply sovereign agents empowered by the expressive potential of aggregated data (Gray et al.).

Data publics are more than just aggregates of individual data points or connections. They are inherently unstable, dynamic (despite static analysis and visualisations), or vibrant, and ephemeral. We emphasise three key elements of active data publics. First, to be more than an aggregate of individual items, a data public needs to be consequential (in Dewey’s sense of issues or problem-oriented). Second, sufficient connection is visible over time. Third, affective or emotional activity is apparent in relation to events that lend coherence to the public and its prevailing sentiment. To these, we add critical attention to the affordising processes – or the deliberate and incidental effects of datafication and analysis, in the capacities for data collection and processing in order to produce particular analytical outcomes, and the data literacies these require. We return to the latter after elaborating on the Save the Palace case….(More)”.

Countries Can Learn from France’s Plan for Public Interest Data and AI

Curated on Posted on by Stefaan Verhulst

Nick Wallace at the Center for Data Innovation: “French President Emmanuel Macron recently endorsed a national AI strategy that includes plans for the French state to make public and private sector datasets available for reuse by others in applications of artificial intelligence (AI) that serve the public interest, such as for healthcare or environmental protection. Although this strategy fails to set out how the French government should promote widespread use of AI throughout the economy, it will nevertheless give a boost to AI in some areas, particularly public services. Furthermore, the plan for promoting the wider reuse of datasets, particularly in areas where the government already calls most of the shots, is a practical idea that other countries should consider as they develop their own comprehensive AI strategies.

The French strategy, drafted by mathematician and Member of Parliament Cédric Villani, calls for legislation to mandate repurposing both public and private sector data, including personal data, to enable public-interest uses of AI by government or others, depending on the sensitivity of the data. For example, public health services could use data generated by Internet of Things (IoT) devices to help doctors better treat and diagnose patients. Researchers could use data captured by motorway CCTV to train driverless cars. Energy distributors could manage peaks and troughs in demand using data from smart meters.

Repurposed data held by private companies could be made publicly available, shared with other companies, or processed securely by the public sector, depending on the extent to which sharing the data presents privacy risks or undermines competition. The report suggests that the government would not require companies to share data publicly when doing so would impact legitimate business interests, nor would it require that any personal data be made public. Instead, Dr. Villani argues that, if wider data sharing would do unreasonable damage to a company’s commercial interests, it may be appropriate to only give public authorities access to the data. But where the stakes are lower, companies could be required to share the data more widely, to maximize reuse. Villani rightly argues that it is virtually impossible to come up with generalizable rules for how data should be shared that would work across all sectors. Instead, he argues for a sector-specific approach to determining how and when data should be shared.

After making the case for state-mandated repurposing of data, the report goes on to highlight four key sectors as priorities: health, transport, the environment, and defense. Since these all have clear implications for the public interest, France can create national laws authorizing extensive repurposing of personal data without violating the General Data Protection Regulation (GDPR) which allows national laws that permit the repurposing of personal data where it serves the public interest. The French strategy is the first clear effort by an EU member state to proactively use this clause in aid of national efforts to bolster AI….(More)”.

Most Public Engagement is Worthless

Curated on Posted on by Stefaan Verhulst

Charles Marohn at Strong Towns: “…Our thinking is a byproduct of the questions we ask. …I’m a planner and I’m a policy nerd. I had all the training in how to hold a public meeting and solicit feedback through SWOT (strengths, weaknesses, opportunities, threats) questions. I’ve been taught how to reach out to marginalized groups and make sure they too have a voice in the process. That is, so long as that voice fit into the paradigm of a planner and a policy nerd. Or so long as I could make it fit.

Modern Planner: What percentage of the city budget should we spend on parks?

Steve Jobs: Do you use the park?

Our planning efforts should absolutely be guided by the experiences of real people. But their actions are the data we should be collecting, not their stated preferences. To do the latter is to get comfortable trying to build a better Walkman.  We should be designing the city equivalent of the iPod: something that responds to how real people actually live. It’s a messier and less affirming undertaking.

I’ve come to the point in my life where I think municipal comprehensive planning is worthless. More often than not, it is a mechanism to wrap a veneer of legitimacy around the large policy objectives of influential people. Most cities would be better off putting together a good vision statement and a set of guiding principles for making decisions, then getting on with it.

That is, get on with the hard work of iteratively building a successful city. That work is a simple, four-step process:

  1. Humbly observe where people in the community struggle.
  2. Ask the question: What is the next smallest thing we can do right now to address that struggle?
  3. Do that thing. Do it right now.
  4. Repeat.

It’s challenging to be humble, especially when you are in a position, or are part of a profession, whose internal narrative tells you that you already knowwhat to do. It’s painful to observe, especially when that means confronting messy realities that do not fit with your view of the world. It’s unsatisfying, at times, to try many small things when the “obvious” fix is right there. If only those around you just shared your “courage” to undertake it (of course, with no downside to you if you’re wrong). If only people had the patience to see it through (while they, not you, continue to struggle in the interim).

Yet what if we humbly observe where people in our community struggle—if we use the experiences of others as our data—and we continually take the actions we are capable of taking, right now, to alleviate those struggles? And what if we do this in neighborhood after neighborhood across the entire city, month after month and year after year? If we do that, not only will we make the lowest risk, highest returning public investments it is possible to make, we won’t help but improve people’s lives in the process….(More)”.

To the smart city and beyond? Developing a typology of smart urban innovation

Curated on Posted on by Stefaan Verhulst

Maja Nilssen in Technological Forecasting and Social Change: “The smart city is an increasingly popular topic in urban development, arousing both excitement and skepticism. However, despite increasing enthusiasm regarding the smartness of cities, the concept is still regarded as somewhat evasive. Encouraged by the multifaceted character of the concept, this article examines how we can categorize the different dimensions often included in the smart city concept, and how these dimensions are coupled to innovation. Furthermore, the article examines the implications of the different understandings of the smart city concept for cities’ abilities to be innovative.

Building on existing scholarly contributions on the smartness of cities and innovation literature, the article develops a typology of smart city initiatives based on the extent and types of innovations they involve. The typology is structured as a smart city continuum, comprising four dimensions of innovation: (1) technological, (2) organizational, (3) collaborative, (4) experimental.

The smart city continuum is then utilized to analyze empirical data from a Norwegian urban development project triggered by a critical juncture. The empirical data shows that the case holds elements of different dimensions of the continuum, supporting the need for a typology of smart cities as multifaceted urban innovation. The continuum can be used as an analytical model for different types of smart city initiatives, and thus shed light on what types of innovation are central in the smart city. Consequently, the article offers useful insights for both practitioners and scholars interested in smart city initiatives….(More)”

Programmers need ethics when designing the technologies that influence people’s lives

Curated on Posted on by Stefaan Verhulst

Cherri M. Pancake at The Conversation: “Computing professionals are on the front lines of almost every aspect of the modern world. They’re involved in the response when hackers steal the personal information of hundreds of thousands of people from a large corporation. Their work can protect – or jeopardize – critical infrastructure like electrical grids and transportation lines. And the algorithms they write may determine who gets a job, who is approved for a bank loan or who gets released on bail.

Technological professionals are the first, and last, lines of defense against the misuse of technology. Nobody else understands the systems as well, and nobody else is in a position to protect specific data elements or ensure the connections between one component and another are appropriate, safe and reliable. As the role of computing continues its decades-long expansion in society, computer scientists are central to what happens next.

That’s why the world’s largest organization of computer scientists and engineers, the Association for Computing Machinery, of which I am president, has issued a new code of ethics for computing professionals. And it’s why ACM is taking other steps to help technologists engage with ethical questions….

ACM’s new ethics code has several important differences from the 1992 version. One has to do with unintended consequences. In the 1970s and 1980s, technologists built software or systems whose effects were limited to specific locations or circumstances. But over the past two decades, it has become clear that as technologies evolve, they can be applied in contexts very different from the original intent.

For example, computer vision research has led to ways of creating 3D models of objects – and people – based on 2D images, but it was never intended to be used in conjunction with machine learning in surveillance or drone applications. The old ethics code asked software developers to be sure a program would actually do what they said it would. The new version also exhorts developers to explicitly evaluate their work to identify potentially harmful side effects or potential for misuse.

Another example has to do with human interaction. In 1992, most software was being developed by trained programmers to run operating systems, databases and other basic computing functions. Today, many applications rely on user interfaces to interact directly with a potentially vast number of people. The updated code of ethics includes more detailed considerations about the needs and sensitivities of very diverse potential users – including discussing discrimination, exclusion and harassment….(More)”.