Why Mathematical Rationalists make Irrational Decisions

Geef een reactie

Book Review of: B. Recht (2026). The Irrational Decision. How we gave Computers the Power to Choose for Us, Princeton University Press, Princeton & Oxford, x + 270 pp.

Voor een Nederlandse vertaling klik hier.

Dr G.J. van Bussel

The author

Benjamin Recht is Professor in the Department of Electrical Engineering and Computer Sciences at the University of California, Berkeley. He has received several prizes and awards recognizing the quality and impact of his research and has served on the editorial boards of the Journal of Machine Learning Research and Mathematical Programming. He is co‑founder of the Conference on Learning for Decision and Control. Recht’s research group investigates how to design machine learning systems that remain robust when interacting with dynamic and uncertain environments. He develops reliable benchmarks and baselines for the assessment of performance of machine learning technology. He has demonstrated, for example, that standard machine learning theory often mischaracterizes the behaviour of deep neural networks and that a number of results in personalized medicine overstate their practical utility. His research is strengthened by close collaborations with scholars in applied domains such as computational imaging and robotics. 1

The key thesis

The key thesis is introduced and presented in the first chapter. Recht prefers the term ‘mathematical rationality’ to describe the narrow, statistical conception of reason that inspired the desire to build computers, shaped how they would eventually operate, and constrained the kinds of problems they would be considered well suited to solve. The Irrational Decision traces how, in the 1940s, mathematicians converged on a distinctive definition of rationality in which every decision is cast as a statistical problem of risk. This notion of rationality is heavily mathematical, and it has proved attractive to hyper‑online rationalists, public intellectuals, and billionaire technologists who (quite irrational) take their own success as evidence of uniquely clear thinking rather than luck or structural advantage.

Lees verder →

Share This:

Archiving Machines

Geef een reactie

Review article of: A. Acker (2025). Archiving Machines. Technology and the Future of Memory Institutions, MIT Press, Cambridge (Ms)-London, 248 pp. To download as open source: https://direct.mit.edu/books/oa-monograph/6055/Archiving-MachinesFrom-Punch-Cards-to-Platforms.

Voor een Nederlandse vertaling klik hier.

Dr G.J. van Bussel

The author

Amelia Acker is an associate professor in the School of Communication and Information at Rutgers University. Her scholarship focuses on the generation, standardization, and preservation of digital information. Her research centres on three pillars: data technologies (the tools used to generate and capture information), digital archives (the management of digital data), and information in­frastruc­tures (the systems that facilitate long-term preservation and serve as repositories for cultural memory). Prior to her appointment at Rutgers, she served as Director of the Critical Data Studies Lab at the University of Texas at Austin. She led interdisciplinary research initiatives concerning the societal impact of ubiquitous mobile technology, public and academic understanding of data systems, and frameworks that enable information to remain searchable and functional over time. 1

The key thesis

Archiving Machines. From Punch Cards to Platforms offers an historical overview of storage technologies and processes, tracing their evolution from the data banks of the 1960s to the emergence of contemporary digital platforms. At the heart of Acker’s argument is the assertion that storage systems function as ‘archiving machines’ that actively shape memory, regulate access, and distribute power. Grounded in Science and Technology Studies and Infrastructure Theory, but also using insights from Media Studies, she contends that these machines define the boundaries of cultural memory, determine its control, and define its organization.

As she clarifies, these machines are ‘not about traditional archives found in institutions but rather about how accessing knowledge has transformed as computational pro­cesses for storing and managing data developed, thus putting pressure on concepts of archives, archive, and archiving. In other words, this book explores how ‘archive’ became a verb with the rise of data-­driven recordkeeping to unpack the stakes of archiving machines’ (p. 21). Her work emphasizes the systemic impact of digital and platform-based infrastructures on collective memory, in particular the ways in which automation and corporate platforms have assumed roles once reserved for archival institutions and professional archivists.

She states ‘how firms who play an infrastructural role in data archives have become ­ today’s archons – ­ those that govern and assert power through the functional sovereignty of data archives’ (p. 11).Acker demonstrates how storage technologies have not only redefined what is preserved but have also reshaped conceptions of memory, ownership, and institutional accountability. Her thesis – that archiving machines shape memory and produce distinct regimes of access and data sovereignty – provides historical context for the challenges in digital stewardship that she identifies in the book. 2

Lees verder →

Share This:

Slow Burn. The Everyday consequences of Climate Change.

Geef een reactie

Book review of: R. Jisung Park (2024). Slow Burn. The Hidden Costs of a Warming World, Princeton University Press, 336 pp. For a Dutch translation of this review see here.

Dr. G.J. van Bussel

R. Jisung Park is an environmental and labour economist and Assistant Professor at the University of Pennsylvania. He holds concurrent appointments within the School of Social Policy and Practice and the Wharton School’s Department of Business Economics and Public Policy. His expertise lies in identifying the impact of environmental factors on economic outcomes. Before joining the University of Pennsylvania, he was a faculty member at the University of California, Los Angeles, and held a postdoctoral fellowship at Harvard University. Employing data analysis, experimental methods, and economic theory, his research examines environmental change, its consequences for economic systems, and the effectiveness of policy responses. Specifically, he investigates issues such as the effects of heat on workers, the impacts of natural disasters on human capital, and the mechanisms through which workers and firms adapt to a changing environment.

In his book Slow Burn. The Hidden Costs of a Warming World, he offers a fresh and compelling perspective on the impact of climate change. The book’s central thesis is that conventional assessments underestimate the true economic, social, and biological costs of a warming planet because they focus on dramatic, discrete catastrophes, such as exceptional hurricanes or acute droughts (‘fast burns’, like, as Park explains, the eruption of Mount Vesuvius in 79 CE).

Park shifts the focus towards the subtle, chronic and everyday consequences that are already occurring, such as the erosion of agricultural crops, the gradual extension of pollen seasons, the strain on public health from prolonged heatwaves, and the combined economic impact of reduced labour productivity and disrupted supply chains. The ‘slow burn’ is the opposite of the ‘fast burn’: the cumulative, incremental and often invisible erosion of human capabilities and economic productivity caused by a gradually changing environment. Park compares this with the gradual decline of the Roman Empire over several centuries.

Lees verder →

Share This:

More and More and More

Geef een reactie

Book review of J.B. Fressoz (2024). More and More and More. An All-Consuming History of Energy, Penguin Random House, 400 pp. For a Dutch translation of this review see here.

Dr. G.J. van Bussel

The focus on energy, and the attached corporate and governmental agendas, has exerted a large influence upon the social sciences. Although political ecology maintains a stance of hostility towards the claims of sustainable development, [1] it has acquiesced to the proposition of an energy transition and the concepts associated with it: ‘green’, ‘clean’, and ‘renewable’ energy. [2] This narrative of transition combines the profit-driven aspirations of green capitalism with a pacification of existing climate anxieties. It is for this reason that reading Jean-Baptiste Fressoz’s new book More and More and More. An All-Consuming History of Energy, is a ‘must’.

Fressoz, a historian at the French National Centre for Scientific Research (CNRS) and Professor at the École des Ponts ParisTech, examines the historical dynamics between humanity and its environment. His work challenges conventional narratives of industrial progress and ecological awareness. In his monograph, Happy Apocalypse. A History of Technological Risk (2024), Fressoz puts forward the provocative idea that environmental and industrial regulations, introduced from the early nineteenth century onwards, were rarely designed to reduce environmental degradation or protect workers. Instead, he argues that their main purpose was to legalize these dangers, thereby creating a legal framework that allowed industrialists to pursue profit. Fressoz suggests that this process of legal normalization systematically externalized the costs of industrialization onto society and the natural world.

Lees verder →

Share This:

A Catalyst for Urgent Dialogue and Reform

1 reactie

Book Review of C. Piller (2025). Doctored. Fraud, Arrogance, and Tragedy in the Quest to Cure Alzheimer’s, Atria, One Signal Publishers, New York, Amsterdam, London, 352 pp.

Dr. G.J. van Bussel

Charles Piller is an investigative journalist renowned for exposing scientific misconduct, fraud, and ethical violations in biomedical research and public health. Writing for publications like Science, STAT News, and the Los Angeles Times, he examines how financial incentives, and institutional pressures compromise scientific integrity, covering topics from the reproducibility crisis to genetic engineering ethics and whistleblower challenges. Through meticulous document analysis, insider interviews, and critical scrutiny, Piller reveals the complex interplay between science, policy, and corporate influence, advocating for transparency and accountability while highlighting how systemic failures undermine public trust in scientific institutions. Piller’s investigations have triggered significant debate, paper retractions, and underscored journalism’s crucial role in safeguarding scientific integrity. From 2022 to 2024, he focused extensively on neuroscience fraud, particularly in Alzheimer’s research, publishing groundbreaking articles in Science that exposed data and image manipulation in clinical trials and revealed how Cassava Biosciences and researcher Hou-Yan Wang had manipulated data regarding the Alzheimer’s drug simufilam. [i] His February 2025 book Doctored. Fraud, Arrogance, and Tragedy in the Quest to Cure Alzheimer’s synthesizes these individual cases into a comprehensive critique of the global Alzheimer’s research ecosystem, exposing how institutional policies, scientific misconduct, and distorted priorities have undermined the quest for effective treatments while leaving patients vulnerable to ineffective or dangerous experimental therapies.

In the Introduction, Piller articulates his central thesis: the dominance of the amyloid cascade hypothesis (the theory that Alzheimer’s disease stems from amyloid plaque accumulation) has stifled alternative research while fostering hype, wasted funding, and questionable breakthroughs. He is not the first to do so. [ii] Using Leqembi (a drug introduced in late 2022 that promised to be the long-awaited Alzheimer’s breakthrough) as his prime example [iii], Piller argues: “It was the latest example of the exaggeration, hype, and sheer fakery and fraud that has characterized Alzheimer’s research for decades. … I had to show why billions of dollars in spending by governments, pharma companies, and philanthropies had done little for desperate patients. For decades, proponents of the dominant amyloid hypothesis have sidelined, starved for resources, and even bullied rebels behind other promising notions of how to treat Alzheimer’s … Alzheimer’s research has offered endless opportunities for advancement and riches to corporate shysters and ruthlessly ambitious scientists who cut corners or engage in brazen deception. I set out to unmask decades of arrogance, greed, fabulism, and error that have emptied research coffers and littered the drug development landscape with failure after failure.” (pp. x-xi.) This narrative summary, while effectively encapsulating his book’s scope, opens him to criticism that the work could have been much shorter.

Lees verder →

Share This:

AI as a Mirror

Geef een reactie

Review of S. Vallor (2024). The AI Mirror. How to Reclaim Our Humanity in an Age of Machine Thinking. Oxford University Press, New York, 263 p.

Dr. G.J. van Bussel

Shannon Vallor is a professor of the ethics of data and artificial intelligence (AI) at the University of Edinburgh. In this role, she explores how new technologies such as AI, robotics, and data science are reshaping human moral character, cognitive habits, and social practices. As one of today’s leading philosophers of technology, she is a key figure in contemporary debates about humanity’s technological future. Her 2016 book, Technology and the Virtues, is an attempt to rethink virtue ethics in the context of technological upheaval and breakthroughs. In it, she presents ‘technomoral virtue ethics’, a framework that critiques how technological breakthroughs erode traditional moral capacities, while also proposing strategies for cultivating new virtues adapted to our era of technological disruption. [i] Her latest book, The AI Mirror (2024), builds on this research and focuses on three interconnected themes: the moral agency of AI systems; the effects of algorithms on human self-understanding; and the conditions needed for responsible innovation. In The AI Mirror, Vallor provides a moral and philosophical analysis of AI and its role in shaping knowledge, values, and morality.

Nearly a year after publication, The AI Mirror remains relevant, its title perfectly expressing Vallor’s central thesis: what we call ‘AI’ is not intelligence but a reflection — one that mimics human cognition while fundamentally lacking understanding. The mirror analogy proves very powerful, exposing how these systems merely process and refract historical data imbued with human conceptions (and misconceptions) including conscious or unconscious biases and misrepresentations of reality. [ii] AI serves as a mirror that exposes and intensifies societal biases, aspirations, and moral contradictions. Vallor argues that AI not only mirrors but recursively influences human cognition.

Employing the mirror metaphor, she warns that without deliberate human intervention, AI will perpetuate biases, distortions, and blind spots rather than fostering meaningful progress. These systems ‘aren’t designed to be accurate — they are designed to sound accurate’ (p. 121.) This distinction is vital, highlighting how large language models trade in plausibility rather than truth, and how their outputs, despite their fluency, remain cognitively hollow. Vallor asserts that the self-image projected by AI is deceptive and that an uncritical fascination with AI poses significant dangers. ‘It is their power to induce in us a type of self-forgetting — a selective amnesia that loosens our grip on our own human agency and clouds our self-knowledge. It is an illusion that can ensnare even the most technologically adept among us’ (p. 2.)

Lees verder →

Share This:

Ideological Platforms

Geef een reactie

Book review of: M. Pepi (2025). Against Platforms: Surviving Digital Utopia, Melville House, Brooklyn-London, 224 p.

Dr. G.J. van Bussel

After the publications of Srnicek’s Platform Capitalism (2016) and Zuboff’s Surveillance Capitalism (2019), there has been an exponential increase in the number of critiques of large technological firms and their platforms. [i] Concurrently, however, digital platforms have consolidated their influence over global societies. In Against Platforms, Mike Pepi, advances the argument that such platforms are not impartial, but rather ideological instruments that exert a significant influence on social, artistic, and political life. Pepi contests the techno-utopian perspective that platforms serve as instruments for ‘good’, for democratization, and equality. According to Pepi, digital platforms exhibit an extractive character, a phenomenon that has the effect of eroding autonomy, under the guise of promoting connectivity.

Digital platforms were initially presented as a ‘utopia’, an open space where the free flow of information would ensure a democratic and egalitarian society. Michael Pepi has expressed the opinion that this idea is misguided, at least in its current form. It is:

‘an ideology run amok. Techno-utopianism—the idea that technology, and technology alone, will create a more egalitarian, democratic society—has been around since we have had tools to make labor easier. But this utopia, like all utopias, doesn’t really exist. Despite this misguided attempt, techno-utopianism has effectively synthesized with two other powerful ideologies—techno-determinism and free-market capitalism—to create what many refer to as platform capitalism.’ (p. 2)

Lees verder →

Share This:

The Decay of Culture?

Geef een reactie

Review of C. Rosen (2024). The Extinction of Experience. Being Human in a Disembodied World, W.W. Norton & Company, New York, 272 p.

Have you ever read a book that gave rise to almost as many notes as in the book itself? This is one of them. To be interested in a new approach and to end up very disappointed by the way the book is structured, by its anecdotal character without evidence telling a different story than the book’s title suggests, by the many contradictions, the sloppy research, the repetitive lamentations about disappearing phenomena, the extreme interpretations and one-sidedness of the narrative, and the fact that there are no suggestions for solutions.

In The Extinction of Experience, Christine Rosen eloquently attempts to explain how technology is reshaping the human ‘experience’. She wants readers to think about what it means to be human in a world where digital media seem to overwhelm and take full possession of human attention. Rosen is concerned about what it means to be human when ‘experience’ is hijacked by virtual analogues and becomes disembodied and, more worryingly, monetized. Experiences are, according to Rosen, practices that formed us as individuals, groups, and societies.

“Experiences, broadly considered, are the ways we become acquainted with the world. Direct experience is our first teacher. As we learn to navigate the world around us, we attach meaning to what we encounter. These experiences and meanings differ across times and cultures but share common threads. Different people speak different languages and have different social customs, but the fact that we all have languages and customs is a marker of our shared humanity. Certain types of experience — some rooted deeply in our evolutionary history, such as face-to-face interaction and various forms of pleasure-seeking; others more recent and reflective of cultural norms, such as patience and our sense of public space and place — are fading from our lives. Many of these experiences are what, historically, have helped us form and nurture a shared reality as human beings.” (introduction)

Rosen focuses on what digital platforms have given us, but more importantly, what they have suppressed. Four themes are continuously repeated: the commodification of experience, the collapse of public communal places, the dangers of frictionless interaction, and the need to reclaim ‘embodied’ experiences.

Lees verder →

Share This:

Tearing us Apart

Geef een reactie

Book review of: N. Carr (2025). Superbloom. How Technologies of Connection Tear Us Apart, W. W. Norton & Company, 272 pages.

The question of whether the world has become a better place since the arrival of ‘technologies of connection’, as Nicholas Carr calls them, is a complex one. These technologies, which include all forms of hardware and software designed to facilitate connectivity between individuals, such as phones, computers, tablets, internet browsers, social media platforms, and large language models, have become an integral part of our lives. However, the potential implications of these technologies for democracy, as well as the dissemination of misinformation, hallucinations, and deepfakes that threaten societal norms and values, ethical considerations, and democratic plurality, appear to contradict the naïve idealism expressed in Silicon Valley’s manipulative marketing. However, the advent of GPS has facilitated global navigation, enabling individuals to communicate with others, even in remote locations, and to access and stream a vast array of audio and video content. It is a mixed bag, with many advantages for the individual and a lot of disadvantages for society. [i]

Nicholas Carr has published extensively about business, information technology and culture, and has expressed highly critical views on the effects of these on individual users and society at large. His book The Shallows. What the Internet Is Doing to Our Brains was a finalist for the 2011 Pulitzer Prize in General Nonfiction. In this provocative book, Carr advanced the argument that Google and the internet are ‘making us stupid.’ It established Carr as a leading commentator on the Internet and its societal consequences. In my review of the book (in Dutch), I stated the following:

Lees verder →

Share This:

To Control History

Geef een reactie

Book review of: J. Stanley (2024). Erasing History. How Fascists Rewrite the Past to Control the Future, Atria/One Signal Publishers, New York, 256 pp.

In this book, the American philosopher and Yale University professor, Jason Stanley, explores the practice of governments manipulating the past to shape a future that aligns with their desired vision. This is a very common practice, even in democratic societies. Uncomfortable truths are erased, certain books are banned from schools, and accepted history is replaced with myths, thereby justifying the power of the governing elites. To succeed, it is necessary for these elites to ascertain that the populace is unaware of what really happened. [i] Especially in authoritarian regimes, it is important to control narratives about the past.

Studying history is dangerous. The act of examining the past from multiple vantage points gives rise to a societal depiction that is characterized by historical plurality, diversity, and openness to reinterpretation. These reinterpretations, in turn, give rise to novel perspectives that serve to augment the historical plurality that is already extant. Within democratic societies, this plurality serves to nurture principles of equality, collective reflection, and the acceptance of a multitude of historical narratives. Each of these narratives offers a distinct perspective on truth and reality, thereby contributing to a rich tapestry of historical understanding. Conversely, within authoritarian states, history is often simplified and mythologized into a single, overarching narrative, serving to justify the supremacy of specific social groups or segments.

Lees verder →

Share This:

Overstating a Case

Geef een reactie

Review of: J. Haidt (2024). The Anxious Generation: How the Great Rewiring of Childhood is Causing an Epidemic of Mental Illness, Penguin Press, 400 pp.

I have read the book on two occasions: first when it was first published, and most recently in the last two weeks. It is a disturbing, thought-provoking, controversial and misleading book.

The book is (and continues to be) a bestseller. It has received a great deal of media attention, and it has been translated into many languages. It examines the rise in mental health issues among the generation born after 1995. Haidt’s central thesis is what he calls ‘the great rewiring of childhood’ that fundamentally changed the experiences of young people and that led to an epidemic rise of anxiety, depression, and other mental health problems. His basic premise is twofold: first, that the introduction of smartphones and the subsequent widespread use of social media has transformed a ‘play-based’ childhood into a ‘phone-based’ one, leading to social and sleep deprivation, attention fragmentation, and addiction, and, second, that ‘parental overprotection’ in the real world is matched with ‘parental underproduction’ in the virtual world, leading to youths being confronted with the negatives of social media, like porn, cyber bullying, dangerous challenges as well as ideal types that can never be met. Haidt contends that all of this led to a collective ‘rewiring’ of the young mind, resulting in increased rates of anxiety, depression, self-harm, and (thoughts of) suicide.

Haidt’s critique focuses on social media platforms, or social networking sites, which he argues are characterized by one-to-many and many-to-one communication. This communication predominantly focuses on the ‘best sides’ (or idealized sides) of user lives to groups of unknown others. Haidt expresses particular concern regarding the use of such platforms by young people (defined as individuals under the age of 18) and asserts that the purported benefits of social media, such as enhanced information management and reputation building, are of negligible relevance to this demographic. Instead, he contends that younger users, particularly those between the ages of 10 and 14, primarily employ social media for the purpose of entertainment. The fact that their brains (particularly the regions involved in regulating behaviour) are not yet fully developed means that this aspect of social media platforms has the potential to become addictive, due to the difficulty of discontinuing usage. Haidt posits that contemporary social media platforms are problematic due to several factors, including the prevalence of endless scrolling, push notifications, the emphasis on enticing video content, and personalization. In my opinion, notwithstanding the tendency to reach premature conclusions given the data and analysis presented, Haidt’s thesis does possess a degree of credibility. Social media is a far more compelling and engaging form of entertainment than reading books or watching television. It is a well-documented fact that young people never dedicated 30-35 hours per week to reading or watching television.

Lees verder →

Share This:

The Unaccountability Machine

Geef een reactie

Book review of: D. Davies (2024). The Unaccountability Machine. Why Big Systems Make Terrible Decisions — and How the World Lost Its Mind, Profile Books, 304 pp.

A passenger is refused permission to board an aircraft and makes a plea to a sympathetic flight attendant. Regrettably, the passenger’s grievance pertains to a company policy, and the flight attendant is unable to assist. This scenario exemplifies a scenario in which the onus of decision-making is entrusted not to an individual, but rather to a set of predetermined guidelines.

‘The unsettling thing about this conversation is that you progressively realise that the human being you are speaking to is only allowed to follow a set of processes and rules that pass on decisions made at a higher level of the corporate hierarchy. It’s often a frustrating experience; you want to get angry, but you can’t really blame the person you’re talking to. Somehow, the airline has constructed a state of affairs where it can speak to you with the anonymous voice of an amorphous corporation, but you have to talk back to it as if it were a person like yourself.’ (p. 16)

Davies poses the question of why errors and crises never appear to be attributable to any specific individual or entity. The blame is often attributed to ’the system’, ’the computer’, ’the algorithm’, ’the AI’, or similar entities. Davies terms this phenomenon an ‘accountability sink’, a structural element that absorbs or obscures the consequences of a decision, thereby preventing direct accountability.

The phenomenon of accountability sinks, as postulated by Davies, serves to impede individuals from both willing and unwillingly making or changing decisions and, consequently, being held directly accountable for them.

Lees verder →

Share This:

The escalating water use of AI

1 reactie

Despite efficiency gains in other areas, the exponentially growing demand for AI has led to an increase in the water footprint.

Driven in part by the growth in AI, Google’s scope-1 onsite water consumption (cooling) in 2022 increased by 20% compared to 2021, and Microsoft saw a 34% increase over the same period. The majority of water consumption for server cooling is derived from potable sources.

Additionally, Scope-2 offsite water consumption (electricity generation) is experiencing growth because of AI’s enormous use of computing power. Scope-3 water consumption within the AI supply chain is also mounting. For instance, the fabrication of a single microchip necessitates approximately 2,200 gallons of ultra-pure water. The staggering demand for AI processing necessitates the utilization of thousands of these chips.

The escalating water usage associated with AI is a matter of significant concern. We need to address water scarcity, a pressing global concern exacerbated by a rapidly expanding population, dwindling water resources, and ageing water infrastructures. This concern extends beyond the mere magnitude of AI models’ water usage, underscoring the imperative for developers to proactively address the shared global challenge of water scarcity.

Shaolei Ren elucidated this challenge in: ‘How much water does AI consume?’. See here.

I have earlier mentioned this issue in ‘The ecological footprint of Artificial Intelligence’.

Published in a LinkedIn post, December 2024.

Share This:

The data’s afterlife

Geef een reactie

Have you ever thought about your data’s afterlife? The digital footprints that we leave behind after our demise are so extensive that our identities can persist beyond our mortal existence.

The advent of AI has enabled the possibility to interact with departed loved ones. A young man is still able to compete with his father in an old Xbox racing game, utilizing his father’s saved games. The ‘digital afterlife industry’ provides a posthumous delivery of video, audio, and written messages. The use of AI enables the replication of language patterns and personality traits through the analysis of extensive personal data.

In The Afterlife of Data. What Happens to Your Information When You Die and Why You Should Care, Carl Öhman, an assistant professor at Uppsala University, addresses the issue of data ownership and the decision-making process regarding the fate of our digital assets after our demise. Who assumes control will effectively control future access to this digital past. This information (or substantial portions thereof) will persist, barring the dissolution of Meta (Facebook) and Alphabet (Google) or a conflict that results in the destruction of the data centres used to store these data.

The afterlife of data raises a multitude of ethical, economic, and legal inquiries that are currently unaddressed. The concept of a legal right to privacy does not extend to the deceased. Secrets that were previously thought to be confined to the grave can, in some cases, have a detrimental impact on the reputation of the deceased. In theory, individuals have the option to delete their data, at least any materials they do not wish to be accessible to the general public. Data, however, will endure. Some parents have attempted to access and process the data of their deceased child, but have been unsuccessful. The courts have determined that it is not within their purview to make such decisions, leaving it to the platform’s owners. Neither the EU nor the US has legislation in place governing post-mortem privacy.

The data is held by many organizations, the majority of which are unknown. Most are unwilling to delete these data. Öhman expresses concerns that profit-driven businesses may lack the reliability required to act as guardians of digital heritage. He questions whether governments or NGOs would be more suitable curators, but is unsure and cites George Orwell’s observation that ‘who controls the past, controls the future.’ However, these data could also prove beneficial to historians, policymakers, and society at large. Roy Rosenzweig observed that the data provide researchers with an ‘essentially complete historical record’.

The implications of all of this are significant and warrant further reflection. ‘The Afterlife of Data’ is an insightful book that offers a compelling perspective on the potential implications of data in the afterlife. It is highly recommended reading for those interested in this topic.

C. Öhman (2024). The Afterlife of Data: What Happens to Your Information When You Die and Why You Should Care, University of Chicago Press, 200 pp.

Share This:

De Nabije Toekomst van Archivering

Geef een reactie

Nieuwe publicatie!

‘De toeslagenaffaire is nog niet voorbij of de malaise van de WIA buitelt eroverheen. De chaos bij de overheid groeit zo uit tot ongekende proporties waardoor de democratie verder onder druk komt te staan. Niet door het gewoonlijke en bijbehorende politiek gekonkel in de Tweede Kamer, maar omdat het ambtelijk apparaat zich naar binnen heeft gekeerd en het contact met burgers en bedrijven heeft verloren, uitzonderingen daargelaten.’

Dus worden de oude systemen en bestaande technieken gebruikt om de problemen aan te pakken. In een zoektocht naar oplossingen lijkt geld niet het probleem: er zijn miljarden uitgegeven zonder resultaat. Maar over de echte aanpak van de toeslagenaffaire is nooit goed nagedacht.

Welke feiten, middelen of regelingenwaren nodig voor effectieve actie? Waarom werden toeslagen eigenlijk stopgezet bij vermeende fraude? Is het probleem überhaupt oplosbaar? Handmatige informatieverzameling zonder bruikbaarheid voor gedupeerden, medewerkers, belangenbehartigers en machines om intelligent over regelingen heen te redeneren, leveren geen resultaten op. Technologieën van 20 jaar geleden, toen state-of-the-art, werken echt niet meer en alleen nieuwe middelen, nieuwe techniek inzetten, is het recept voor een volgende affaire.

Techniek zonder een goed concept en een heldere architectuur geeft slechts een illusie van hoop. De status quo hindert de oplossing omdat de communicatie al in de basis is verstoord. Het is onbeheersbaar complex geworden. De oplossing ligt in openheid, we moeten als overheid anders leren denken: transparant zijn en fouten erkennen.

Citatie: Geert-Jan van Bussel en Hans van Bommel (2024). ‘De nabije toekomst van archivering’, jrg, 78, november 2024, nr. 39, pp. 24-25.

Artikel Open op OrdeDownload

Share This:

AI’s Power Demand

Geef een reactie

AI results in a large increase of data center power demand, and does have a large effect on natural resources.

In 2018, OpenAI concluded that the computing power required to train a large AI model had doubled every 3.5 months from 2012 onwards. The accuracy of results and time efficiency that can be achieved by harnessing the computing power of a vast number of computers in data centres necessitates a considerable amount of electricity. A significant proportion of data centres globally continue to rely, to some extent, on fossil fuels, resulting in a notable surge in CO₂ emissions.

In 2020, researchers at the University of Massachusetts conducted an analysis of several natural language processing (NLP) models and determined that the energy expenditure associated with training a single model resulted in CO2 emissions of approximately 300,000 kg on average (equivalent to 125 return flights from New York to Beijing). The training of ChatGPT-3 has been found to require the consumption of 1.3 gigawatt hours of electricity, resulting in the generation of 550,000 kg of CO2. It is estimated by Bloomberg that the energy consumption necessary for training is only 40% of that required for operational purposes. Moreover, the training process necessitates the consumption of approximately 700,000 litres of water for the purpose of computer cooling. This quantity of water is equivalent to that which would be required by a nuclear power plant cooling tower.

In 2023, data centres operated by Google extracted a total of 24 billion litres of water from the environment. This represents a 14% increase compared to figures recorded in the previous year. In 2022, 20 billion litres of water were employed for the purpose of cooling. Two-thirds of this quantity was comprised of potable water.

Furthermore, data from Microsoft’s facilities indicate a 34% increase in cooling water consumption during the same period. In 2024, Microsoft’s CO₂ emissions were 30% higher than in 2020, while Google’s emissions increased by 48% over the past five years.

See this blogpost by GoldmanSachs

For literature used to compose this post, see here.

Publishd in a LinkedIn-post, november 2024.

Share This:

Co-Intelligence

Geef een reactie

I have recently read Ethan Mollick’s book, Co-intelligence. Living and working with AI. The book is well written and an easy and enjoyable read. The author explores the potential of conversational AI (LLMs) in the context of personal and professional realms.

As the book makes clear, LLMs are adept at creative work and at identifying unanticipated correlations. It can falter in elementary intellectual pursuits. It is a statistical model of language that generates a plausible imitation of human responses by calculating the probability of clusters of words. The responses are devoid of cognitive processes, knowledge, or logical validation. They are the result of calculations regarding the optimal sequence of words. The provided response does not need to be accurate or true.

Mollick’s book presents four guiding principles for the practical and ethical use of LLMs:
[1] Use LLMs to investigate and delineate the boundaries of their functionalities.
[2] Engage in active monitoring of the responses provided, ensuring that they are accurate, true, and free from hallucinations.
[3] Treat conversational AI as a person. Critique its answers, clarify where necessary, reformulate questions, and offer feedback.
[4] Treat every LLM as if it is the least sophisticated model you ‘will ever use,’ recognizing that new iterations or models may offer improvements.

What struck me is the banality of conversational AI. It will be unknown whether the words were written by conscious minds or AIs and have ‘feeling’. Mollick refers to this as a ‘crisis of meaning’, but a crisis of language seems more accurate.

Mollick emphasizes the concept of ‘co-intelligence’, humans and AI working together to enhance productivity, creativity and decision-making. He proposes a symbiotic relationship in which AI tools enhance human capabilities. The book is full of anecdotes, examples, and texts that seek to demonstrate the veracity of that proposition. There is, however, no theoretical basis to support this assertion.

In contradiction, an article in Nature Human Behaviour in October 2024 found that, on average, human–AI combinations performed significantly worse than the best of humans or AI alone. See: M. Vaccaro, A. Almaatouq, and T. Malone (2024). ‘When combinations of humans and AI are useful: A systematic review and meta-analysis’, Nature Human Behaviour (2024). https://doi.org/10.1038/s41562-024-02024-1.

The book’s implications for the use of conversational AI in an educational context are interesting. This is in line with Mollick’s educational experience at the Wharton School of Business. He requires the use of such tools by his students. Allegedly, a teacher has the potential to significantly enhance the quality of learning outcomes.

The book’s lack of academic depth and its anecdotal nature make it an excellent introduction for those with limited knowledge of AI. I can recommend reading it, especially for students.

E. Mollick (2024). Co-Intelligence: Living and Working with AI, Penguin Portfolio, 256 pp.

Published as e LinkedIn post here.

Share This:

In de bureaulade

Geef een reactie

Erna Havinga publiceerde onlangs een blogpost waarin ze ingaat op het verschijnsel dat bestuurders compliance issues negeren, zelfs als er auditrapporten liggen die wijzen op misstanden. Ze richt zich vooral op ‘informatiebeveiliging en privacy’.

Het verschijnsel is mij niet vreemd: ik weet dat vele advies- en auditrapporten over problemen met de informatiehuishouding in bureaulades gelegd worden en dat de aanbevelingen slechts marginaal worden opgevolgd. Redenen daarvoor zijn divers, maar een ervan is dat de noodzaak niet zo gevoeld wordt en dat het in ieder geval geen prioriteit heeft.

Die onachtzaamheid speelt al veel langer als het gaat om informatie, informatiemanagement en informatiebeveiliging. Er is genoeg wetenschappelijke literatuur beschikbaar waaruit blijkt dat door organisatie- en medewerkersgedrag vele mooie informatie(beleids)plannen onderuit worden gehaald. En daar doen managers en bestuurders aan mee. Ze hebben de mond vol over ‘informatie op orde’, maar ze kijken het liefst weg als het aan ‘de orde’ komt. Of, ook vaak toegepast, gaan weer een nieuw systeem implementeren dat ‘alle problemen oplost’. Het nemen van ingrijpende beslissingen kan daardoor weer worden uitgesteld en er hoeven ook geen ingrijpende maatregelen genomen te worden om voor ‘gewenst gedrag’ te zorgen.

Wat me wel van het hart moet: ‘hoezo privacy’? Daar gaat het toch helemaal niet over? De AVG spreekt over de ‘inbreuk op de vertrouwelijkheid, integriteit of beschikbaarheid van persoonsgegevens’. Het woord privacy komt helemaal niet voor in de AVG, en dat is niet zo vreemd omdat privacy niet iets is wat beschermd kan worden door een wet.

Privacy by design en Privacy Impact Analyses bestaan niet. Het zijn volgens de AVG ‘gegevensbescherming door ontwerp’ en ‘effectbeoordeling inzake gegevensbescherming’. Dat is iets heel anders en veel concreter dan ‘privacy’. Het gaat dus eigenlijk om informatiebeheer, -management en -beveiliging.

Informatie is iets waar menselijk gedrag en organisatiecultuur moeite mee hebben. Er wordt veel geroepen, maar weinig geregeld en ingegrepen. Hard roepen dat informatie belangrijk is, is makkelijker dan dat echt goed organiseren in een complexe omgeving.

Het is niet zo vreemd dat negatieve auditresultaten niet leiden tot actie. Ingrijpen om ‘gewenst gedrag’ te stimuleren is immers ‘moeilijk’, zeker in de informele organisatieculturen waar vele organisaties prat op gaan.

Voor gedrag in organisaties:

G.J. van Bussel, A Sound of Silence. Organizational Behaviour and Enterprise Information Management, Van Bussel Document Services, Helmond 2020.
Online bron: hier en hier.

E, Havinga (2024), ‘Rapporteren zonder resultaat; de frustratie van elke FG en CISO’. IB&P Blog. Online bron: hier.

Share This:

Harari’s Nexus

Geef een reactie

I am not the first to comment on Yuval Harari’s new book Nexus. A Brief History of Information Networks from the Stone Age. The reception has been mixed. I was curious to see if Harari would offer more than ‘big ideas’ and ‘big history’ that doesn’t require archival sources and close reading before drawing conclusions. In Sapiens, Harari proposed big ideas with abandon, such as gossip as the basis of the ‘cognitive revolution’ and the agricultural revolution as the ‘biggest fraud’ in history. Both may be big ideas, but they are also untrue. His generalizations paint beautiful, well-written and easy to understand anecdotes, but most of them are just that: anecdotes, mostly unconfirmed, unproven and without context.

Nexus is a disappointing read. The book could have been shorter than the 528 pages Harari expects us to read. There is much that is interesting, even brilliant. But most are anecdotes without context or truisms presented as original ideas. Harari sets out to write about information, switches to writing about information media and, in the end, about artificial intelligence, which is neither.

But perhaps these stories and truisms are at the heart of the book? Early in the book, Harari debunks the ‘naive view of information’, the idea that gathering more information leads to truth, wisdom and power. According to Harari, there is no correlation between information, truth and power. Information networks are not about truth, they are about stories. Think of religions. They are successful because they have a large information network and they tell the same story. While his point about religions may be true, information networks are not just about stories. There are proven links between control of information, control of (imagined) truth and power.

The second half of the book is about computers and artificial intelligence, which Harari believes could lead to Armageddon. Most of his concerns seem reasonable, such as generative AIs poisoning online discussions. He becomes a prophet when he discusses how ‘AI overlords’ might create a new pandemic, a new kind of money, and a flood of fake news or incitement to riot on information networks. Fake news and incitement are already true, and we don’t need advanced ‘AI overlords’ for that. And a solution? Regulation and ‘institutions with strong self-correcting mechanisms’. Answering with a truism to fight the ‘end of days’? Such a solution will not prevent a situation in which, as he writes in the prologue, ‘we might find ourselves nowhere at all, as Earth becomes inhospitable for human life’. He could have done better.

Nexus lacks focus. Anecdotes from all over space and time are juxtaposed and accompanied by truisms, such as, at the end of chapter 11: ‘The only constant in history is change’. Which is true, of course, but I did not need Harari to know that.

Perhaps I am a naive information theorist. And maybe that is why I do not recommend this book for reading.

Y.N. Harari (2024). Nexus. A Brief History of Information Networks from the Stone Age, Random House, New York, 528 pp.

Published in a post on LinkedIn, September 2024.

Share This:

AI and digital preservation

Geef een reactie

Interesting paper (february 2024) about the role of AI for the preservation of archives.

According to the abstract:
‘This paper delves into the historical evolution of preservation methods driven by technological advancements as, throughout history, libraries, archives, and museums have grappled with the challenge of preserving historical collections, while many of the traditional preservation methods are costly and involve a lot of manual (human) effort. AI being the catalyst for transformation could change this reality and perhaps redefine the process of preservation; thus, this paper explores the emerging trend of incorporating AI technology into preservation practices and provides predictions regarding the transformative role of Artificial Intelligence in preservation for the future. With that in mind, this paper addresses the following questions: could AI be what changes or creates a paradigm shift in how preservation is done?; and could it be the thing that will change the way history is safeguarded?’

I don’t really know if it would be a paradigm shift or that it changes the way archives are safeguarded.

That AI might redefine the process of preservation: I don’t doubt that. The process of preservation can be automated and the information needed to do so is available. So, it is quite possible that an Artificial Narrow Intelligence would be of enormous help in processing al those routine tasks now done by people.

But: it would mean indexing all known data about the process (processes) of preservation and meticulously designing an AI driven process, with several human controls.

Citation:
Z. Teel (2024). ‘Artificial Intelligence’s Role in Digitally Preserving Historic Archives’, Preservation, Digital Technology & Culture, Vol 53, NO. 1. Online source. For download here.

Published in a LinkedIn post in september 2024.

Share This: