Paper by the Centre for Information Policy Leadership: “Contemporary everyday life is increasingly permeated by digital information, whether by creating, consuming or depending on it. Most of our professional and private lives now rely to a large degree on digital interactions. As a result, access to and the use of data, and in particular personal data, are key elements and drivers of the digital economy and society. This has brought us to a significant inflection point on the issue of legitimising the processing of personal data in the wide range of contexts that are essential to our data-driven, AI-enabled digital products and services. The time has come to seriously re-consider the status of consent as a privileged legal basis and to consider alternatives that are better suited for a wide range of essential data processing contexts. The most prominent among these alternatives are the “legitimate interest” and “contractual necessity” legal bases, which have found an equivalent in a number of jurisdictions. One example is Singapore, where revisions to their data protection framework include a legitimate interest exemption…(More)”.
Article by Emilio Mariscal: “…After some exploration, I came up with an idea: what if we could export chat conversations and extract the location data along with the associated messages? The solution would involve a straightforward application where users can upload their exported chats and instantly generate a map displaying all shared locations and messages. No business accounts or complex integrations would be required—just a simple, ready-to-use tool from day one.
ChatMap —chatmap.hotosm.org — is a straightforward and simple mapping solution that leverages WhatsApp, an application used by 2.78 billion people worldwide. Its simplicity and accessibility make it an effective tool for communities with limited technical knowledge. And it even works offline! as it relies on the GPS signal for location, sending all data with the phone to gather connectivity.
This solution provides complete independence, as it does not require users to adopt a technology that depends on third-party maintenance. It’s a simple data flow with an equally straightforward script that can be improved by anyone interested on GitHub.
We’re already using it! Recently, as part of a community mapping project to assess the risks in the slopes of Comuna 8 in Medellín, an area vulnerable to repeated flooding, a group of students and local collectives collaborated with the Humanitarian OpenStreetMap (HOT) to map areas affected by landslides and other disaster impacts. This initiative facilitated the identification and characterization of settlements, supporting humanitarian aid efforts.
Photo by Daniela Arbeláez Suárez (source: WhatsApp)
As shown in the picture, the community explored the area on foot, using their phones to take photos and notes, and shared them along with the location. It was incredibly simple!
The data gathered during this activity was transformed 20 minutes later (once getting access to a WIFI network) into a map, which was then uploaded to our online platform powered by uMap (umap.hotosm.org)…(More)”.
Article by Stefaan Verhulst and Adam Zable: “Non-Traditional Data (NTD): “data that is digitally captured (e.g. mobile phone records), mediated (e.g. social media), or observed (e.g. satellite imagery), using new instrumentation mechanisms, often privately held.”
Digitalization and the resulting datafication have introduced a new category of data that, when re-used responsibly, can complement traditional data in addressing public interest questions—from public health to environmental conservation. Unlocking these often privately held datasets through data collaboratives is a key focus of what we have called The Third Wave of Open Data.
To help bridge this gap, we have curated below recent examples of the use of NTD for research and decision-making that were published the past few months. They are organized into five categories:
Article by Stefaan Verhulst: “It was the best of times, it was the worst of times… It was the spring of hope, it was the winter of despair.”–Charles Dickens, A Tale of Two Cities
Charles Dickens’s famous line captures the contradictions of the present moment in the world of data. On the one hand, data has become central to addressing humanity’s most pressing challenges — climate change, healthcare, economic development, public policy, and scientific discovery. On the other hand, despite the unprecedented quantity of data being generated, significant obstacles remain to accessing and reusing it. As our digital ecosystems evolve, including the rapid advances in artificial intelligence, we find ourselves both on the verge of a golden era of open data and at risk of slipping deeper into a restrictive “data winter.”
A Tale of Two Cities by Charles Dickens (1902)
These two realities are concurrent: the challenges posed by growing restrictions on data reuse, and the countervailing potential brought by advancements in privacy-enhancing technologies (PETs), synthetic data, and data commons approaches. It argues that while current trends toward closed data ecosystems threaten innovation, new technologies and frameworks could lead to a “Fourth Wave of Open Data,” potentially ushering in a new era of data accessibility and collaboration…(More)” (First Published in Industry Data for Society Partnership’s (IDSP) 2024 Year in Review).
NDI Report: “The dawn of a new space age is upon us, marked by unprecedented engagement from both state and private actors. Driven by technological innovations such as reusable rockets and miniaturized satellites, this era presents a double-edged sword for global democracy. On one side, democratized access to space offers powerful tools for enhancing civic processes. Satellite technology now enables real-time election monitoring, improved communication in remote areas, and more effective public infrastructure planning. It also equips democratic actors with means to document human rights abuses and circumvent authoritarian internet restrictions.
However, the accessibility of these technologies also raises significant concerns. The potential for privacy infringements and misuse by authoritarian regimes or malicious actors casts a shadow over these advancements.
This report discusses the opportunities and risks that space and satellite technologies pose to democracy, human rights, and civic processes globally. It examines the current regulatory and normative frameworks governing space activities and highlights key considerations for stakeholders navigating this increasingly competitive domain.
It is essential that the global democracy community be familiar with emerging trends in space and satellite technology and their implications for the future. Failure to do so will leave the community unprepared to harness the opportunities or address the challenges that space capabilities present. It would also cede influence over the development of global norms and standards in this arena to states and private sector interests alone and, in turn, ensure those standards are not rooted in democratic norms and human rights, but rather in principles such as state sovereignty and profit maximization…(More)”.
Article by Nicola Jones: “The Internet is a vast ocean of human knowledge, but it isn’t infinite. And artificial intelligence (AI) researchers have nearly sucked it dry.
The past decade of explosive improvement in AI has been driven in large part by making neural networks bigger and training them on ever-more data. This scaling has proved surprisingly effective at making large language models (LLMs) — such as those that power the chatbot ChatGPT — both more capable of replicating conversational language and of developing emergent properties such as reasoning. But some specialists say that we are now approaching the limits of scaling. That’s in part because of the ballooning energy requirements for computing. But it’s also because LLM developers are running out of the conventional data sets used to train their models.
A prominent study1 made headlines this year by putting a number on this problem: researchers at Epoch AI, a virtual research institute, projected that, by around 2028, the typical size of data set used to train an AI model will reach the same size as the total estimated stock of public online text. In other words, AI is likely to run out of training data in about four years’ time (see ‘Running out of data’). At the same time, data owners — such as newspaper publishers — are starting to crack down on how their content can be used, tightening access even more. That’s causing a crisis in the size of the ‘data commons’, says Shayne Longpre, an AI researcher at the Massachusetts Institute of Technology in Cambridge who leads the Data Provenance Initiative, a grass-roots organization that conducts audits of AI data sets.
The imminent bottleneck in training data could be starting to pinch. “I strongly suspect that’s already happening,” says Longpre…(More)”
However, the index’s success became its downfall. Some governments set up dedicated teams with an explicit goal of improving the country’s performance on the index. If those teams’ activity was solely focussed on positive policy reform, that would be great; unfortunately, in at least some cases, they were simply trying to game the results.
Index ranking optimisation (aka gaming the results)
To give an example of how that could happen, we need to take a brief detour into the world of qualitative indicators. Bear with me. In many indexes grappling with complex topics, there is a perennial problem of data availability. Imagine you want to measure the number of days it takes to set up a new business (this was one of the indicators in Doing Business). You will find that most of the time the data either doesn’t exist or is rarely updated by governments. Instead, put very simplistically, you’d need to ask a few experts or businesses for their views, and use those to create a numerical score for your index.
This is a valid approach, and it’s used in a lot of studies. Take Transparency International’s long-running Corruption Perceptions Index (CPI). Transparency International goes to great lengths to use robust and comparable data across countries, but measuring actual corruption is not viable — for obvious reasons. So the CPI does something different, and the clue is in the name: it measures people’s perceptions of corruption. It asks local businesses and experts whether they think there’s much bribery, nepotism and other forms of corruption in their country. This foundational input is then bolstered with other data points. The data doesn’t aim to measure corruption; instead, it’s about assessing which countries are more, or less, corrupt.
This technique can work well, but it got a bit shaky as Doing Business’s fame grew. Some governments that were anxious to move up the rankings started urging the World Bank to tweak the methodology used to assess their ratings, or to use the views of specific experts. The analysts responsible for assessing a country’s scores and data points were put under significant pressure, often facing strong criticism from governments that didn’t agree with their assessments. In the end, an internal review showed that a number of countries’ scores had been improperly manipulated…The criticism must have stung, because the team behind the World Bank’s newBusiness Ready report has spent three years trying to address those issues. The new methodology handbook lands with a thump at 704 pages…(More)”.
UNESCO Paper: “The deployment of advanced Artificial Intelligence (AI) models, particularly generative AI, has sparked discussions regarding the creation and use of synthetic content – i.e. AI-generated or modified outputs, including text, images, sounds, and combinations thereof – and its impact on individuals, societies, and economies. This note explores the different ways in which synthetic content can be generated and used and proposes a taxonomy that encompasses synthetic media and deepfakes, among others. The taxonomy aims to systematize key characteristics, enhancing understanding and informing policy discussions. Key findings highlight both the potential benefits and concerns associated with synthetic content in fields like data analytics, environmental sustainability, education, creativity, and mis/disinformation and point to the need to frame them ethically, in line with the principles and values of UNESCO Recommendation on the Ethics of Artificial Intelligence. Finally, the note brings to the fore critical questions that policymakers and experts alike need to address to ensure that the development of AI technologies aligns with human rights, human dignity, and fundamental freedoms…(More)”.
Paper by Aaron Martin and Bryce Newell: “Public and scholarly interest in the related concepts of synthetic data and synthetic media has exploded in recent years. From issues raised by the generation of synthetic datasets to train machine learning models to the public-facing, consumer availability of artificial intelligence (AI) powered image manipulation and creation apps and the associated increase in synthetic (or “deepfake”) media, these technologies have shifted from being niche curiosities of the computer science community to become topics of significant public, corporate, and regulatory import. They are emblematic of a “data-generation revolution” (Gal and Lynskey 2024: 1091) that is already raising pressing questions for the academic surveillance studies community. Within surveillance studies scholarship, Fussey (2022: 348) has argued that synthetic media is one of several “issues of urgent societal and planetary concern” and that it has “arguably never been more important” for surveillance studies “researchers to understand these dynamics and complex processes, evidence their implications, and translate esoteric knowledge to produce meaningful analysis.” Yet, while fields adjacent to surveillance studies have begun to explore the ethical risks of synthetic data, we currently perceive a lack of attention to the surveillance implications of synthetic data and synthetic media in published literature within our field. In response, this Dialogue is designed to help promote thinking and discussion about the links and disconnections between synthetic data, synthetic media, and surveillance…(More)”
Evidence Brief by NSW Government: “Social licence, otherwise referred to as social licence to operate, refers to an approval or consensus from the society members or the community for the users, either as a public or private enterprise or individual, to use their health data as desired or accepted under certain conditions. Social licence is a dynamic and fluid concept and is subject to change over time often influenced by societal and contextual factors. The social licence is usually indicated through ongoing engagement and negotiations with the public and is not a contract with strict terms and conditions. It is, rather, a moral and ethical responsibility assumed by the data users based on trust and legitimacy, It supplements the techno-legal mechanisms to regulate the use of data. For example, through public engagement, certain values and principles can emerge as pertinent to public support for using their data. Similarly, the public may view certain activities relating to their data use as acceptable and beneficial, implying their permission for certain activities or usecase scenarios. Internationally, although not always explicitly referred to as a social licence, the most common approach to establishing public trust and support and identifying common grounds or agreements on acceptable practices for use of data is through public engagement. Engagement methods and mechanisms for gaining public perspectives vary across countries (Table 1). − Canada – Health Data Research Network Canada reports on social licence for uses of health data, based on deliberative discussions with 20 experienced public and patient advisors. The output is a list of agreements and disagreements on what uses and users of health data have social licence. − New Zealand – In 2022, the Ministry of Health commissioned a survey on public perceptions on use of personal health information. This report identified conditions under which the public supports the re-use of their data…(More)”.
