Stefaan Verhulst
Essay by Joshua Rothman: “Last spring, Daniel Kokotajlo, an A.I.-safety researcher working at OpenAI, quit his job in protest. He’d become convinced that the company wasn’t prepared for the future of its own technology, and wanted to sound the alarm. After a mutual friend connected us, we spoke on the phone. I found Kokotajlo affable, informed, and anxious. Advances in “alignment,” he told me—the suite of techniques used to insure that A.I. acts in accordance with human commands and values—were lagging behind gains in intelligence. Researchers, he said, were hurtling toward the creation of powerful systems they couldn’t control.
Kokotajlo, who had transitioned from a graduate program in philosophy to a career in A.I., explained how he’d educated himself so that he could understand the field. While at OpenAI, part of his job had been to track progress in A.I. so that he could construct timelines predicting when various thresholds of intelligence might be crossed. At one point, after the technology advanced unexpectedly, he’d had to shift his timelines up by decades. In 2021, he’d written a scenario about A.I. titled “What 2026 Looks Like.” Much of what he’d predicted had come to pass before the titular year. He’d concluded that a point of no return, when A.I. might become better than people at almost all important tasks, and be trusted with great power and authority, could arrive in 2027 or sooner. He sounded scared.
Around the same time that Kokotajlo left OpenAI, two computer scientists at Princeton, Sayash Kapoor and Arvind Narayanan, were preparing for the publication of their book, “AI Snake Oil: What Artificial Intelligence Can Do, What It Can’t, and How to Tell the Difference.” In it, Kapoor and Narayanan, who study technology’s integration with society, advanced views that were diametrically opposed to Kokotajlo’s. They argued that many timelines of A.I.’s future were wildly optimistic; that claims about its usefulness were often exaggerated or outright fraudulent; and that, because of the world’s inherent complexity, even powerful A.I. would change it only slowly. They cited many cases in which A.I. systems had been called upon to deliver important judgments—about medical diagnoses, or hiring—and had made rookie mistakes that indicated a fundamental disconnect from reality. The newest systems, they maintained, suffered from the same flaw.Recently, all three researchers have sharpened their views, releasing reports that take their analyses further. The nonprofit AI Futures Project, of which Kokotajlo is the executive director, has published “AI 2027,” a heavily footnoted document, written by Kokotajlo and four other researchers, which works out a chilling scenario in which “superintelligent” A.I. systems either dominate or exterminate the human race by 2030. It’s meant to be taken seriously, as a warning about what might really happen. Meanwhile, Kapoor and Narayanan, in a new paper titled “AI as Normal Technology,” insist that practical obstacles of all kinds—from regulations and professional standards to the simple difficulty of doing physical things in the real world—will slow A.I.’s deployment and limit its transformational potential. While conceding that A.I. may eventually turn out to be a revolutionary technology, on the scale of electricity or the internet, they maintain that it will remain “normal”—that is, controllable through familiar safety measures, such as fail-safes, kill switches, and human supervision—for the foreseeable future. “AI is often analogized to nuclear weapons,” they argue. But “the right analogy is nuclear power,” which has remained mostly manageable and, if anything, may be underutilized for safety reasons.
Briefing paper from the Food Data Collaboration: “…a call to action to collaborate and invest in data infrastructure that will enable shorter, relational, regenerative food supply networks to scale.
These food supply networks play a vital role in achieving a truly sustainable and resilient food system. By embracing data technology that fosters commons ownership models, collaboration and interdependence we can build a more inclusive and dynamic food ecosystem in which collaborative efforts, as opposed to competitive businesses operating in silos, can achieve transformative scale.
Since 2022, the Food Data Collaboration has been exploring the potential for open data standards to enable shorter, relational, regenerative food supply networks to scale and pave the way towards a healthier, more equitable, and more resilient food future. This paper explores the high level rationale for our approach and is essential reading for anyone keen to know more about the project’s aims, achievements and future development…(More)”.
Whitepaper by the Global Government Technology Centre Berlin: “…explores how agentic AI will transform ten functional layers of government and public administration. The Agentic State signifies a fundamental shift in governance, where AI systems can perceive, reason, and act with minimal human intervention to deliver public value. Its impact on key functional layers of government will be as follows…(More)”.
Article by Andreas Pawelke, Basma Albanna and Damiano Cerrone: “Cities around the world face urgent challenges — from climate change impacts to rapid urbanization and infrastructure strain. Municipal leaders struggle with limited budgets, competing priorities, and pressure to show quick results, making traditional approaches to urban transformation increasingly difficult to implement.
Every city, however, has hidden success stories — neighborhoods, initiatives, or communities that are achieving remarkable results despite facing similar challenges as their peers.
These “positive deviants” often remain unrecognized and underutilized, yet they contain the seeds of solutions that are already adapted to local contexts and constraints.
Data-Powered Positive Deviance (DPPD) combines urban data, advanced analytics, and community engagement to systematically uncover these bright spots and amplify their impact. This new approach offers a pathway to urban transformation that is not only evidence-based but also cost-effective and deeply rooted in local realities.
DPPD is particularly valuable in resource-constrained environments, where expensive external solutions often fail to take hold. By starting with what’s already working, cities can make strategic investments that build on existing strengths rather than starting from scratch. Leveraging AI tools that improve community engagement, the approach becomes even more powerful — enabling cities to envision potential futures, and engage citizens in meaningful co-creation…(More)”
Paper by Janet Freilich and W. Nicholson Price II: “A large literature on regulation highlights the many different methods of policy-making: command-and-control rulemaking, informational disclosures, tort liability, taxes, and more. But the literature overlooks a powerful method to achieve policy objectives: data. The state can provide (or suppress) data as a regulatory tool to solve policy problems. For administrations with expansive views of government’s purpose, government-provided data can serve as infrastructure for innovation and push innovation in socially desirable directions; for administrations with deregulatory ambitions, suppressing or choosing not to collect data can reduce regulatory power or serve as a back-door mechanism to subvert statutory or common law rules. Government-provided data is particularly powerful for data-driven technologies such as AI where it is sometimes more effective than traditional methods of regulation. But government-provided data is a policy tool beyond AI and can influence policy in any field. We illustrate why government-provided data is a compelling tool both for positive regulation and deregulation in contexts ranging from addressing healthcare discrimination, automating legal practice, smart power generation, and others. We then consider objections and limitations to the role of government-provided data as policy instrument, with substantial focus on privacy concerns and the possibility for autocratic abuse.
We build on the broad literature on regulation by introducing data as a regulatory tool. We also join—and diverge from—the growing literature on data by showing that while data can be privately produced purely for private gain, they do not need to be. Rather, government can be deeply involved in the generation and sharing of data, taking a much more publicly oriented view. Ultimately, while government-provided data are not a panacea for either regulatory or data problems, governments should view data provision as an understudied but useful tool in the innovation and governance toolbox…(More)”
Book by Anne Trumbore: “From AI tutors who ensure individualized instruction but cannot do math to free online courses from elite universities that were supposed to democratize higher education, claims that technological innovations will transform education often fall short. Yet, as Anne Trumbore shows in The Teacher in the Machine, the promises of today’s cutting-edge technologies aren’t new. Long before the excitement about the disruptive potential of generative AI–powered tutors and massive open online courses, scholars at Stanford, MIT, and the University of Illinois in the 1960s and 1970s were encouraged by the US government to experiment with computers and artificial intelligence in education. Trumbore argues that the contrast between these two eras of educational technology reveals the changing role of higher education in the United States as it shifted from a public good to a private investment.
Writing from a unique insider’s perspective and drawing on interviews with key figures, historical research, and case studies, Trumbore traces today’s disparate discussions about generative AI, student loan debt, and declining social trust in higher education back to their common origins at a handful of elite universities fifty years ago. Arguing that those early educational experiments have resonance today, Trumbore points the way to a more equitable and collaborative pedagogical future. Her account offers a critical lens on the history of technology in education just as universities and students seek a stronger hand in shaping the future of their institutions…(More)”
Paper by Anu Masso, Anniki Puura, Jevgenia Gerassimenko and Olle Järv: “The European Strategy for Data aims to create a unified environment for accessing, sharing, and reusing data across sectors, institutions, and individuals, with a focus on areas like mobility and smart cities. While significant progress has been made in the technical interoperability and legislative frameworks for data spaces, critical gaps persist in the bottom-up processes, particularly in fostering social collaboration and citizen-driven initiatives. What is often overlooked is the need for effective citizen engagement and collaborative governance models to ensure the long-term viability and inclusivity of these data spaces. In addition, although principles for successful data sharing are well-established in sectors like healthcare, they remain underdeveloped and more challenging to implement in areas such as mobility. This article addresses these gaps by exploring how gamification can drive bottom-up data space formation, engaging citizens in data-sharing and fostering collaboration among private companies, local governments, and academic institutions. Using bicycle usage as an example, it illustrates how gamification can incentivise citizens to share mobility data for social good, promoting more active and sustainable transportation in cities. Drawing on a case study from Tallinn (Estonia), the paper demonstrates how gamification can improve data collection, highlighting the vital role of citizen participation in urban planning. The article emphasises that while technological solutions for data spaces are advancing, understanding collaborative governance models for data sharing remains crucial for ensuring the success of the European Union’s data space agenda and driving sustainable innovation in urban environments…(More)”.
Article by Simon Makin: “In 1785 English philosopher Jeremy Bentham designed the perfect prison: Cells circle a tower from which an unseen guard can observe any inmate at will. As far as a prisoner knows, at any given time, the guard may be watching—or may not be. Inmates have to assume they’re constantly observed and behave accordingly. Welcome to the Panopticon.
Many of us will recognize this feeling of relentless surveillance. Information about who we are, what we do and buy and where we go is increasingly available to completely anonymous third parties. We’re expected to present much of our lives to online audiences and, in some social circles, to share our location with friends. Millions of effectively invisible closed-circuit television (CCTV) cameras and smart doorbells watch us in public, and we know facial recognition with artificial intelligence can put names to faces.
So how does being watched affect us? “It’s one of the first topics to have been studied in psychology,” says Clément Belletier, a psychologist at University of Clermont Auvergne in France. In 1898 psychologist Norman Triplett showed that cyclists raced harder in the presence of others. From the 1970s onward, studies showed how we change our overt behavior when we are watched to manage our reputation and social consequences.
But being watched doesn’t just change our behavior; decades of research show it also infiltrates our mind to impact how we think. And now a new study reveals how being watched affects unconscious processing in our brain. In this era of surveillance, researchers say, the findings raise concerns about our collective mental health…(More)”.
Paper by Amirhosein Shabrang, Mehdi Pourpeikari Heris, and Travis Flohr: “We investigated the spatial shade patterns of trees and buildings on sidewalks and bike lanes in Cambridge, Massachusetts. We used Lidar data and 3D modeling to analyze the spatial and temporal shade distribution across the City. Our analysis shows significant shade variations throughout the City. Western city areas receive more shade from trees, and the eastern regions receive more shade from buildings. The City’s northern areas lack shade, but natural and built sources of shade can improve shade coverage integration. This study’s findings help identify shade coverage gaps, which have implications for urban planning and design for more heat-resilient cities…(More)”
Book edited by Michael Lewis: “The government is a vast, complex system that Americans pay for, rebel against, rely upon, dismiss, and celebrate. It’s also our shared resource for addressing the biggest problems of society. And it’s made up of people, mostly unrecognized and uncelebrated, doing work that can be deeply consequential and beneficial to everyone.
Michael Lewis invited his favorite writers, including Casey Cep, Dave Eggers, John Lanchester, Geraldine Brooks, Sarah Vowell, and W. Kamau Bell, to join him in finding someone doing an interesting job for the government and writing about them. The stories they found are unexpected, riveting, and inspiring, including a former coal miner devoted to making mine roofs less likely to collapse, saving thousands of lives; an IRS agent straight out of a crime thriller; and the manager who made the National Cemetery Administration the best-run organization, public or private, in the entire country. Each essay shines a spotlight on the essential behind-the-scenes work of exemplary federal employees.
Whether they’re digitizing archives, chasing down cybercriminals, or discovering new planets, these public servants are committed to their work and universally reluctant to take credit. Expanding on the Washington Post series, the vivid profiles in Who Is Government? blow up the stereotype of the irrelevant bureaucrat. They show how the essential business of government makes our lives possible, and how much it matters…(More)”.