Stefaan Verhulst

Essay by Grace Huckins: “In 2001, a decade after he invented the World Wide Web, Tim Berners-Lee articulated a new vision. Together with two colleagues, he published an article in the magazine Scientific American in which he outlined what he called the “Semantic Web”: an internet designed to be navigated by autonomous software agents that could address our every need, without requiring us to muck about with browsers and search engines ourselves.

To illustrate what this might look like, Berners-Lee and his colleagues told a story about a pair of siblings who use their agents to automatically research and book physical therapy appointments for their mother, making sure they would be covered by insurance and fit their busy calendars. 

At the turn of the millennium, this vision seemed far-fetched: Google had only been online for a couple of years, and most people were still on dial-up connections. But a quarter-century on, the dream of autonomous software agents that use the internet and carry out tasks for us has finally been realized. New tools such as Meta’s blockbuster Muse agent, OpenAI’s Dots, and Anthropic’s Claude Duet are being given access to people’s email inboxes and bank accounts,  Meanwhile, in a series of worrying incidents, agents created by OpenAI broke free of their sandboxes and caused havoc online.  

Just as the rise of smartphones sparked the development of a more atomized and attention-grabbing internet, the advent of powerful AI systems heralds a new era: We’re facing not the semantic web but the agentic web. This new web is developing out of sight and in ways that might not be entirely within our control. As it grows, the agentic web will inevitably transform the human web—and whether the human web survives this transition is anyone’s guess…(More)”.

Article by Ann Borda and Bastian Greshake Tzovaras: “In this perspectives article, the authors review key factors generating problematic oversight and participatory data governance gaps around AI (artificial intelligence) and machine learning (ML) applications and the role of crowdsourced citizen science in addressing these gaps.  Within this context, there are challenges in a broader AI ecosystem of power asymmetries lacking participative governance of data use by commercial providers and in unregulated commodification of citizen science data. Such challenges potentially can diminish the effectiveness of crowdsourced citizen science approaches in the participatory data governance of AI/ML applications. Notwithstanding, crowdsourced citizen science has a keystone role in its ethical grounding in communities, and this study proposes consideration of crowdsourced citizen science as contributors to public interest policy and alternative models of collective data governance operationalization. These models can potentially support crowdsourced citizen science as co-partners through multistakeholder processes, such as public–private partnerships, across the AI/ML application lifecycle. A normative framework for collective data governance coalesces perspectives on emerging models, such as data commons and data cooperatives, and repositions the notion of the ‘collective’ moving away from a wholly data-centric approach. Future research is required to address specific collective data governance risks, such as recentralizing control despite participatory paradigms. Although these risks might be mitigated in evolving collective data governance partnerships focused on public good, these are yet to be tested more rigorously and, thus far, partnerships may only be practicable if there is a willing alignment and collective demonstration of societal benefits in an equally rapidly emerging monopolistic and solutionism-directed AI/ML ecosystem…(More)”.

Article by John M. Kamensky: “…A new report by the IBM Center for The Business of Government documents that 21^st century citizens are performing a new role as they engage with their government at all levels – they are becoming innovators.  The study’s authors, University of Wisconsin professors Satish and Priya Nambisan, say that a trifecta of austerity, complexity, and digital technology has created this new opportunity for everyday citizens to help “co-create” new solutions for solving civic problems in their communities.

The term “co-creation” refers to the development of new public services by citizens in partnership with government. The authors outline four roles that citizen co-creators often assume: explorer, ideator, designer, and diffuser.  According to the authors:

“As explorer, citizens can identify, discover, and define emerging and existing problems in public services. For example, the New York-based Datakind initiative involves citizen volunteers using their data analysis skills to mine public data in health, education, environ­ment, and more areas to identify important civic issues and problems.

“As ideator, citizens can conceptualize novel solutions to well-defined problems in public services. For example, initiatives such as federal government’s Challenge.gov and OpenIDEO employ online contests and competitions to solicit innovative ideas to solve important civic problems.

“As designer, citizens can design and/or develop implementable solutions to well-defined problems in public services. For example, as part of initiatives such as NYC Big Apps and Apps for Californians, citizens have designed mobile apps to address specific issues such as public parking availability, public transport delays, and more.

“As diffuser, citizens can directly support or facilitate the adoption and diffusion of public service innovations and solutions among well-defined target populations. For example, physicians interacting with peers in dedicated online communities have assisted govern­ment agencies in diffusing health technology innovations.”..(More)”

Article by Dylan Baysa, Melodie A. Lopez & Lana Fred:  “For years, nearly every piece of data the Tucson Indian Center (TIC) collected was designed to answer someone else’s questions. Since 1963, TIC has served the urban Native community in Tucson, Arizona, offering primary care and behavioral health services, diabetes prevention and wellness programs, employment and job training, and emergency housing assistance. Intake forms for those programs, satisfaction surveys, and reporting metrics were all developed to collect data that met federal grant requirements but told us almost nothing about what our community wanted to know about itself.

Eventually, those of us leading and working at TIC said, “Enough.” We began building a different kind of data system—one organized around our community’s priorities. But to understand why this work is both difficult and transformative requires understanding the origins of contemporary data systems that are used to quantify Native people in the United States.

For much of the 19th and 20th centuries, information about Native peoples was collected and controlled primarily by federal agencies, academic institutions, and other outside entities. Rooted in settler colonialism and white supremacy, that control was part of broader efforts to dominate Native land, identity, and governance, and it has left many Native nations dependent on data rarely collected by or for them. Data, including census records and blood-quantum classifications, were used to define Native identity, determine eligibility for land and government benefits, and advance assimilation policies, often with limited Tribal authority over how information was collected, interpreted, shared, or used. In this way, quantification became a tool of administration and control. Research also frequently took a deficit-based approach, emphasizing poverty, health disparities, unemployment, and other challenges while overlooking community strengths, cultural knowledge, Tribal priorities, and the social and historical conditions shaping those outcomes. In doing so, it reinforced stereotypes and informed policy without adequately reflecting Indigenous perspectives. Many of today’s data failures, such as the chronic failure to collect mortality data for Indigenous populations, continue that history through seemingly mundane administrative policy and practice…(More)”.

Paper by Yuhan Li,Nicolò Gozzi,Nicola Perra & Michele Tizzoni: “The concept of fairness has been extensively examined within the domains of Machine Learning and Artificial Intelligence more broadly. It remains, however, largely underexplored in the field of Computational Epidemiology. Considering the substantial influence that epidemic models exert on public health policy, particularly in the context of outbreak preparedness and response, this shortcoming is of great relevance. Here, we propose a mathematical framework, grounded in core principles from Social Epidemiology, for evaluating the fairness of computational epidemic models. We begin by applying our framework to a range of epidemic modeling approaches and simulation scenarios, such as the initial spread of COVID-19 in London, New York, and Santiago de Chile, as well as the 2016 Zika virus outbreak in Colombia, demonstrating its consistent capacity to assess model fairness across diverse disease dynamics. Subsequently, we illustrate how our definition of fairness can be incorporated into the design of immunization strategies as a way to enhance health equity while simultaneously improving the overall effectiveness of such interventions. Overall, our results offer a new systematic methodology for quantifying fairness in computational epidemiology…(More)”.

Article by Kyriaki Papageorgiou: “It is tempting to think of CERN only as a triumph of scale and engineering: a vast complex of accelerators, detectors, and tunnels built to probe the fundamental structure of matter. This image often stands in for “Big Science” more generally, reinforcing the idea that CERN’s primary lesson lies in technical ambition or concentrated resources. But this reading misses what has made CERN historically distinctive and politically durable. From its founding in 1954, in the aftermath of the Second World War, CERN was conceived not simply as a laboratory, but as a political infrastructure for rebuilding European science (and Europe) through cooperation. It was designed to pool resources and research capacity across the borders of its original 12 member states after decades of fragmentation and scientific brain drain. Its founding convention embedded a public, nonmilitary ethos from the outset, explicitly committing the organization to “have no concern with work for military requirements.”

CERN’s governance was treaty-based and intergovernmental, tying scientific collaboration to the shared funding commitments and collective decisionmaking of the CERN Council, through which member states vote to decide on scientific and administrative issues. Precisely because CERN made large-scale collaboration functional, productive, and credible, it has also become one of the most closely studied scientific infrastructures, offering unusually rich accounts of how trust is built in practice…(More)”.

Article by Anna Tumadóttir and Lisa Macpherson: “…This post outlines the forces that are threatening the digital knowledge commons. These include extractive enclosure — the commodification of human knowledge resources as free raw materials by dominant technology actors — and defensive enclosure — restrictions or demands imposed by online rights holders in self-protection against what they perceive as exploitation. We also outline the dynamics that are poisoning the knowledge commons. We show how, combined, these dynamics are creating a new, unsustainable economic bargain for the open web that may benefit large AI model developers and online rights holders while shutting out everybody else. Lastly, we propose a set of principles for developing solutions to address the impact of these forces on the knowledge commons…(More)”.

Paper by Lauren A. Haygood and Todd Halihan: “More than 50% of Oklahoma’s surface water systems are classified as impaired by the US EPA, with limited funds designated toward surface water monitoring. Community science data can contribute to increasing the amount of quality-controlled government water quality reporting. A framework is presented for implementing a community-based approach to generating quality-controlled water field data to support water quality monitoring by local and regional governments. This research focuses on the design and implementation of a water quality community science project in Oklahoma, USA. In this study, community scientists collected water samples and produced over 100 measurements of key water quality parameters, including alkalinity, conductivity, total dissolved solids, temperature, and selected metals (e.g., iron, manganese, mercury, cadmium, and lead). The frequency and resolution of these community-generated data exceed those of conventional state government monitoring, demonstrating the potential for community science to provide cost-effective, high-frequency, and spatially expansive data sets, which fill monitoring gaps. Additionally, community science can act as a method to bridge the gap between stakeholders, scientists, and policy makers. This study highlights the quality-controlled methods, tools, and community-driven efforts shaping water quality research and public engagement in Oklahoma, a step forward in community-driven data acquisition and water policy…(More)”.

Article by Thiago Moraes: “Since the adoption of the EU AI Act, which requires every EU Member State to establish at least one AI regulatory sandbox, such initiatives have been expanding across the EU and the wider European Economic Area (EEA). But how far have these initiatives progressed, and what can we learn from what has been made public so far? Here is a snapshot of the landscape as of October 2026.

The findings offer a snapshot of an evolving landscape. By the date of analysis, publicly available information pointed to completed AI Act’s sandbox pilots in Denmark, Germany, the Netherlands and Spain. The Spanish initiative stands out as a pioneer, with published statistics and reports and practical guidelines to support the AI Act implementation. Estonia, Latvia and Lithuania had launched pilots, while Sweden, Liechtenstein and Norway had publicly announced sandbox coordinators, but the reviewed information confirmed no sandbox pilots.

Table 1 summarises the findings across all 27 EU Member States and the three non-EU EEA countries. The four categories reflect the stage of development based on public available information at the time of the research…(More)”.

Table 1. Publicly documented state of play of AI Act regulatory sandboxes in the EU and EEA (October 2026)

Note: Data compiled from official public sources as of October 2026. 

Paper by Cuiyu Qin, L. G. Pee, Xiaojie Chen, and Xiaoguang Wang: “In the context of artificial intelligence (AI)-driven scientific research, research information practices are undergoing profound developments. This paper employs a mixed-methods approach, combining topic modeling and interpretive content analysis, to explore emerging research information practices associated with the integration of generative AI into research. This study identifies six significant shifts: from keyword-based search to prompt-based retrieval, from human sensemaking to delegated interpretation, from human innovation to algorithmic co-creation, from manual writing to generative authoring, from unregulated AI use to accountable disclosure, and from AI use to AI literacy. It further analyzes the implications of these emerging research information practices for research productivity, cognitive and critical thinking, academic integrity and ethics, and academic fairness. This study enriches existing research on information practices in evolving information environments and provides a practical path for researchers, libraries, publishers, universities, and research institutions to move beyond the simple use of AI toward the development of AI literacy…(More)”.