Stefaan Verhulst

Policy Brief by the ARC Centre of Excellence for Automated Decision-Making and Society (ADM+S) and the Centre of Artificial Intelligence and Digital Ethics (CAIDE): “…outlines two possible models for the design of an effective and systemic statutory Digital Duty of Care.

• A risk-based model, which would require digital platforms to assess and take reasonable steps to prevent and mitigate the risk of harm arising from the design, operation and use of their services.
• An outcomes-based model, which would require platforms to actively enable a safe, inclusive, healthy and rights-respecting digital environment in the long term.

Strengthening accountability and transparency

The researchers argue that the effectiveness of any Digital Duty of Care will depend on robust accountability frameworks that govern how platforms operate, curate content for, and facilitate engagement between users.

They note that existing transparency measures, which often rely on aggregated reporting, are not sufficient to explain how algorithmic systems shape users’ online experiences.

Instead, they call for new legal and technical infrastructure to enable users, regulators and civil society to meaningfully monitor and observe platform behaviour.

Call for a national platform observatory

A key recommendation of the brief is the establishment of a dedicated national platform observatory, with the mandate and resourcing to track how algorithmic systems target and curate content for Australians.

The observatory would also be responsible for collecting and analysing the information needed to assess whether platforms are complying with the proposed Digital Duty of Care.

Researchers argue that such infrastructure is essential to ensure the transparency, accountability and effectiveness of the new regulatory framework, and to support long-term public trust in digital platforms..(More)”.

Article by Rebecca Winthrop: “…Brainstorming is the work that’s fundamental to writing. As a researcher studying A.I.’s effects on education, I have concluded that these tools only superficially improve writing. The bigger and more alarming impact they have is to constrict our full range of thoughts and our ability to generate original and useful ideas — what we call creative thinking. This seems to be especially true for students. A.I.’s smooth sentences, elegant transitions and rich vocabulary give the illusion of expansive creativity and individuality. But the underlying ideas often converge into a few homogenized categories.

The erosion of creative thinking means young people will struggle to navigate uncertainty. Workers will strain to adapt to a shifting labor market. And society will miss out on the new ideas that can solve complex problems and enhance lives.

For the past eight years, the Georgetown University neuroscientist Adam Green has been leading a national research team tracking the range of novel ideas that college-bound high school students present in their application essays, before and after the introduction of ChatGPT. In one study, he and his team examined personal statements from more than 370,000 students, and found that after ChatGPT became available, their essays suddenly used diverse and colorful language, but lacked truly creative ideas. And the linguistic coverup worked; post-ChatGPT essays were rated as more “creative” by human judges, even if the substance of the essays trod familiar territory…(More)”.

Article by Jennifer Gibson and Kaitlin Thaney: “The global research enterprise relies on information infrastructure to power scientific discovery, medical breakthroughs, and evidence-based policymaking. But the data repositories, digital asset management services, and preservation systems that ensure research data remains open and accessible are often overlooked—until they disappear. Many of these tools and services are vulnerable to policy changes and funding cuts. Over the last 25 years, nearly 200 research data repositories have shut down permanently; more than half of those closures have happened since 2018.

Each closure represents lost knowledge and leads to broken links, bad citations, and a general inability to utilize and verify scientific findings. For example, without funding from the National Oceanic and Atmospheric Administration, the Alaska Earthquake Center has ceased providing real-time seismic data to inform tsunami warnings for the whole US West Coast. On topics as disparate as Gulf War illness or natural selection, when a repository goes dark, it can affect individuals or even entire research disciplines.

The data repositories, digital asset management services, and preservation systems that ensure research data remains open and accessible are often overlooked—until they disappear.

And because repositories and other open science infrastructures are commonly designed to support transboundary research, their collapse can have compounding global effects. In early 2025, the United States Agency for International Development (USAID) suspended access to the Demographic and Health Surveys (DHS) Program databases, a repository containing decades of population, health, HIV, and nutrition data from more than 90 countries. Almost overnight, researchers in Malawi lost access to critical data informing antiretroviral therapy programs serving roughly one million HIV-positive patients; researchers in Nigeria had nowhere to store new data designed to identify causes of maternal deaths; and the release of complete data from a 2023–2024 key indicators survey in the Democratic Republic of the Congo was delayed for months. After USAID was dismantled in the first half of 2025, an emergency grant from the Gates Foundation restored access to existing DHS data and selected surveys. But this three-year support has not returned the program to its prior scale, leaving 23 countries with surveys still incomplete or unanalyzed…(More)”.

This study by CanTrust Hosting Co-operative and Hypha Worker Co-operative: “…examines whether a genuinely co-operative alternative is feasible: one that preserves data sovereignty, democratic governance, and environmental integrity without asking organisations to accept inferior tools or prohibitive costs. Our report covers:

• Market analysis,
• Technical feasibility (including original research on energy consumption),
• Risk assessment,
• And financial viability.

What we found was that such an alternative is both technically achievable, and clearly needed…(More)”.

Book by Sarah O’Connor: “A tsunami of change, we are told, is sweeping the economy as robots and AI threaten to take over tasks done by humans. But while we worry that we’re robotizing our work, what if the real risk is that we’re robotizing ourselves?

When prize-winning Financial Times journalist Sarah O’Connor set out to investigate what was happening on the front lines of technological change, she found people who weren’t losing their jobs to machines, but who felt they were losing something else instead. From translators forced to edit AI output to university graduates interviewed by software and warehouse workers surrounded by robots, she heard stories of work becoming lonelier, less creative, less human.

But O’Connor also found hopeful stories of jobs being made better, safer and more enjoyable – where workers haven’t rejected the new tools, but instead have learned to control them. Exploring questions of power, design, institutions and ideas, her reporting shows that the way technology changes the world of work is not pre-determined, but must be contested and shaped by all of us.

Inspired by stories from nineteenth-century English cotton mills to twenty-first century Swedish mines, We Are Not Machines reveals how we can fight for work which is more respectful of our limits, and more worthy of our minds…(More)”.

Article by Sarah O’Connor: “Arguments about the past are often used as proxies for arguments about the future. It is no surprise, then, that a long-running debate about the Industrial Revolution has flared up as we begin another phase of rapid technological change.

The argument concerns whether the industrial revolution was good or bad for workers in the short run (and, by extension, which the AI revolution will be). The discourse in tech circles can be boiled down to this: “Relax: the industrial revolution led to higher real wages and more jobs”. “But don’t you know about ‘Engels’ pause’? Between 1790 and 1840, profits rose but real wages barely budged.” “Ah, but don’t you know that a different measure of real wages tells a different story?” And so on.

I find this argument perplexing — not because there are no lessons to be learnt from the industrial revolution, but because I don’t think these databases are the right place to look for them.

For one thing, the data on that era is patchy and unreliable. For another, the industrial revolution in Britain took place against a very different institutional backdrop. There was no universal suffrage, no legal trade unions and no modern welfare state. Indeed, you could argue these were eventual social responses to the industrial revolution. It is hard to see why we should expect the wage-setting dynamics of the past (even if we could agree on what they actually were) to repeat themselves today.

But most importantly, these quantitative metrics do not capture how profoundly the industrial revolution changed the nature of work for many people, in ways both good and bad. As the historian EP Thompson puts it in The Making of the English Working Class, “some of the most bitter conflicts of these years turned on issues which are not encompassed by cost-of-living series”: health, working hours, child labour, security and independence…(More)”.

Paper by Sara Thabit, Till Degkwitz, Mahardika Fadmastuti and Luca Mora: “Technology decentralisation is increasingly proposed as a key feature to build more trustworthy, accessible, and innovative digital public infrastructure, yet there is limited empirical knowledge of the actual benefits that such decentralised approaches would create from a public sector perspective. Furthermore, existing literature often relies on a linear dichotomy between data supply and demand that fails to capture the complexity of decentralised data ecosystems. This paper addresses these gaps by adopting an assemblage thinking perspective to conceptualise data platforms as complex socio-technical arrangements, and by developing a public values framework to broaden the understanding of the various outcome that data platforms can create. We apply this approach to an exploratory case study of Hamburg’s Urban Data Platform (UDP). Our findings demonstrate that public value creation is not determined by technical decentralisation alone but by specific architecture-governance configurations. We illustrate that both decentral and central practices can co-occur within the same system, where the role of a leading orchestrator is crucial to drive public value creation…(More)”.

Open Consultation by the Minister of State for Digital Government and Data:  “People generate large volumes of personal data through everyday activities, yet they currently derive limited benefit from it. Instead, this data is held by data controllers, creating a clear power imbalance between individuals and those organisations that collect and control their data.

Data intermediaries offer a way to rebalance this relationship. They offer the potential to empower individuals to take control of their own data, operating as third parties to enable those individuals to better access, share and manage their personal data. The types of data intermediary are varied, but they generally allow users to either regain control of who can access their data, or to share their data on their own terms. By revolutionising where and by whom data is held, intermediaries can unlock new, unrealised benefits from people’s data, ranging from innovative personalised AI services to groundbreaking new research enabled fully by the user’s informed consent.  

Although personal data is constantly produced, its value is still overwhelmingly captured by traditional data controllers rather than by the individuals who generate it. While people have rights as data subjects to access their data, these rights are often under-utilised. Our call for evidence last year found that people’s awareness of what they can do with their data through third parties is limited.

There is a huge opportunity for the UK to use data more strategically, to unlock stronger competition across markets, helping to stimulate innovation and deliver sustained economic growth. The government recognises that intermediaries can play a vital role in unlocking competitive data‑driven markets and supporting innovation, productivity and growth, and is committed to creating the conditions for the easy and secure sharing and reuse of high‑quality data that intermediaries enable. This is a nascent and rapidly developing area of activity in the UK and globally, offering significant potential to enhance outcomes for people, businesses and the wider economy.

Responses to our call for evidence last year indicated there are three things that need to be achieved for intermediaries to be able to function better in the UK: legal ambiguities need to be addressed; data controllers must be confident providing people’s data to another party; and user awareness of intermediaries and their potential value needs to grow. Currently, the barriers around these three areas are limiting the uptake and growth of the sector. Responses suggest these barriers appear interrelated and mutually reinforcing, leading to a sector with potential that is yet to be fully realised.

If the barriers to the sector are addressed, the potential is huge. What if, instead of a data controller typically deciding how your data is used, you were able to exercise greater control over your data and confidently determine who has access to it? Or if you could donate it to research projects to enable new discoveries? Or if it were powering personalised, innovative new AI services that could securely combine your data from multiple sources to provide you with new insights or suggestions? Data intermediaries are essential to unlocking this vision…(More)”.

Article by Kinling Lo: “For decades, the pilgrimage route for ambitious tech founders, investors, and engineers led to Silicon Valley. Now, a growing number of them are flying to Shanghai, Hangzhou, or Shenzhen instead. China is seeing a surge in a new kind of tech tourism where visitors pay up to $9,000 for curated tours of electric-vehicle factories, robotaxis, and artificial intelligence and robotics companies. The trend is partially triggered by viral videos of China’s dancing humanoid robots and flying cars, creating a sense that the country may be moving faster than the West in key emerging technologies. As China-U.S. tech competition intensifies, these trips are also becoming a means to explore the next investment opportunity and tech breakthrough.

“There’s a fear-of-missing-out dynamic at play: the sense that China’s tech ecosystem has reached a level of sophistication where not seeing it firsthand puts you at an informational disadvantage relative to competitors who have,” Shaoyu Yuan, an international relations adjunct professor at New York University who specializes in China’s soft power, told Rest of World…(More)”.

Paper by Lorenzo Gabrielli, Patrizia Sulis, Sara Thabit and Marco Minghini: “Open, non-governmental building datasets have become increasingly important for urban analysis, exposure modelling, and policy support. Despite their growing use, little is known about the consistency, completeness, and comparability of the semantic information they provide at a continental scale. This study presents the first systematic comparison of the semantic attributes of six major pan-European open building datasets—OpenStreetMap, EUBUCCO, Microsoft Global ML Building Footprints, Overture Maps, GHS-OBAT, and the Digital Building Stock Model (DBSM)—using the 27 EU Member States as a common reference area. Five key semantic attributes (height, typology, building age, number of floors, and building material) were harmonised and analysed in terms of completeness and value distributions across countries and degrees of urbanisation. The workflow combines API-based data ingestion, distributed geospatial processing, and high-performance computing to handle around 1.250 billion building footprints. Results reveal pronounced heterogeneity in semantic content across datasets. Remote-sensing-derived products (GHS-OBAT and DBSM) exhibit the highest levels of attribute completeness for height, typology, and building age, but rely on aggregated or coarse semantic representations. In contrast, community-driven and conflated datasets (OpenStreetMap and Overture Maps) provide richer and more detailed semantic schemas, albeit with low and spatially uneven completeness. Completeness patterns vary substantially across countries and urbanisation classes, and high completeness values often mask limited semantic informativeness due to the prevalence of unknown or aggregated attribute values. Overall, the findings demonstrate that no single dataset is universally optimal regarding consistency and completeness of building footprints’ semantic attributes. Nonetheless, the paper provides practical guidance for selecting suitable data sources depending on spatial scale, attribute requirements, and analytical objectives…(More)”.