net.wars

The sovereignty paradox

The year since the 2025 Computers, Privacy, and Data Protection has made Europe more distinct as an entity. Two years ago, we were being chastised for paying insufficient attention to structural corporate power. At this conference last year, we were warned about “simplification”, since published as the Digital Omnibus that threatens to undo some aspects of data protection and other privacy rights.

This year, we heard a lot about “European values”. Invoked as a metric: does “simplification” measure up or is it a sign of weakening? Invoked as a frame to guide developing the digital euro. Invoked as a reason why digital sovereignty is increasingly essential. In November 2025, 23 European countries signed a declaration embracing the latter as a policy.

And yet, although Plixavra Vogiatzoglou introduced a panel discussion by calling digital sovereignty “urgent because of Trump’s trade wars”, she then said it was unrealistic.

There’s a lot to like in the 2025 declaration, which lists as principles open source solutions, the principle of common assets, competitive markets, and democracy. The problem, as Vogiatzoglou said, is that Europe doesn’t make all the necessary components for building its own bottom-to-top stack. The investment required is likely to favor the wealthiest countries, deepening the imbalances that already favor US technology companies.

This is the “sovereignty paradox,” as Zuzanna Warso called it. Replicating the current infrastructure with little change other than a different owner isn’t the right goal – as someone said later, who wants a European Palantir? Warso favors redesigning the technical ecosystem to foster the “digital commons” – and being honest about the tradeoffs.

The same theme reappeared in a discussion of agentic assistants: they will ultimately sit on infrastructure belonging to the same few hyperscalers. Frederika Kaltheuner posited three scenarios: full vertical integration (like Google), integrated models and software (Anthropic and Claude Code), or open source and smaller models, which she thought was Europe’s only opportunity for sovereignty.

All of this seems set to get worse with agentic AI, which, Apple’s Gary Davis said, will allow agents onto all our devices capable of listening, observing, inferring, and acting across apps. What controls do we want? I personally want the control of barring this proposed technological future from my life, but how many of us will have that choice? Davis also noted the European Commission’s release, a few weeks ago, of proposals for requiring Google to allow competing AI services onto Android. If they follow through, he said, it will allow a large-scale privacy and security experiment on European users. While Davis’s employer has its own rasons for opposing this, he has a point. This is untried technology controlled by a handful of companies that could give them overwhelming power over individuals.

Among other new threats to privacy was eye tracking, a constant reality in games played with virtual reality headsets that could easily spread more widely via augmented reality smartglasses. The only way for Europe to counter this, Michael Raschke said, is to create large market-leading companies to act as gatekeepers to intermediate to meet European expectations of security and privacy.

A discussion of the digital euro had this same backdrop: part of the point is to reclaim some of the payments business from US giants Visa and Mastercard. Although, that’s over-simplified: the plans include offline and online versions of the digital euro which do different things. The offline version is meant as a digital reinterpretation of cash that allows anonymous person-to-person payments. The online version is…well, it’s hard to distinguish it from a bank transfer, except that “central bank digital currency” makes stuffy old banks sound kind of cool? Or it did when “crypto” was new and hot. The British equivalent, the digital pound, is in the design phase.

Those thoughts made Leon Schumacher‘s spirited intervention satisfying: where, he asked, was future-proofing against quantum computing, or accommodation for agentic AI, which is expected to underpin…well, no one knows quite how much in transactions, but they’re willing to guess. A trillion dollars, says McKinsey; up to $17.5 trillion, thinks Deloitte, $190 to $385 billion, per Morgan Stanley. Weirdly they all agree on *when*: by 2030. The digital euro is intended to arrive in 2029.

Meanwhile, there is simplification, which has few fans in the privacy world. As Orla Lynskey noted, simplification doesn’t *have* to mean deregulation – however, the European Commission’s proposals reduce rights, facilitate more data processing and *don’t* simplify. Plus, she added, changes of this magnitude require more time for thought and care.

Even business folk present, such as Spanish company founder Alicia Asín Pérez, thought deregulation was less important than many other constraints on business on her list. “My concern is, who area we deregulating for? What will be the consequences?” she said.

What most people favored instead is less fragmentation, and enforcement of the laws we have, a long-running theme at this event.

The former MEP Sophie in t’ Veld had a different take on European values. “We are obsessed with the US,” she said. “But it’s happening right here.” She called out government leaders for ignoring orders from the European Court of Justice, courts for rubber-stamping requests to target journalists with spyware, and called governments not using spyware complicit by their silence.

Illustrations: EU flag (via Wikimedia).

Wendy M. Grossman is an award-winning journalist. Her Web site has an extensive archive of her books, articles, and music, and an archive of earlier columns in this series. She is a contributing editor for the Plutopia News Network podcast. Follow on Mastodon or Bluesky.

First, do no harm

In the midst of the recent Labour leadership turbulence, on Wednesday May 13 Wes Streeting, who would resign from his position as the UK’s health minister a day later, published the Health bill. Among its provisions are the latest in a long line of attempts to centralize and exploit the data collected by the National Health Service. medConfidental provides a useful summary.

Two main aims have kept reappearing over the years. One is to make the health service more efficient and streamline patient interactions. Probably everyone supports this – until they read the details. Even before the bill was published, media reported the government plans to give each of us a single patient record. In the announcement, Streeting talks about the frustration of having to repeat your medical history to every new physician. True; on the other hand, rehashing the medical history is an opportunity to correct errors and misperceptions. This is where patients need choice and control.

The second recurring theme is using patient data to fuel research. Most people support that, too, as long as the data isn’t handed off to commercial companies to exploit for profit. Antecedents here include 2014’s care.data and a its revival in 2021. On his web page, the late Ross Anderson, who extensively researched the security of clinical information systems, documented examples going back to the mid-1990s.

The key complaint about single patient records is the rupture of the individual trust relationship between individuals and their doctors’ surgeries. Traditionally, the GP is the gatekeeper to the rest of the health service. GPs refer you to specialist consultants, provide continuity, and keep your notes. The single care record turns all this into a centralized database under the control of the health minister with many downsides that may not be immediately apparent, particularly to administrators focused on efficiency rather than patient care. Like Facebook, it can be impossible to wholly opt out even if you don’t use the service because others’ records may carve out your shadow.

In March, the worst happened to show the risks: Hannah Devlin and Tom Burgis reported at the Guardian that the data held by UK Biobank turned up for sale on the Chinese site Alibaba. Since its founding in 2003, Biobank has collected detailed longitudinal health records from more than 500,000 volunteers for the purpose of research. The issue seems to have been code and data researchers stored in repositories such as Github, sharing that is often now required by funders. The data was quickly removed, but uncertainty remains.

Even before that, medConfidential warned that pandemic hospital data the government gave to Biobank could be reidentified, and posed risks to health privacy generally. In addition, medConfidential warns that Biobank’s failure to protect its data is systemic and cultural. No one denies Biobank’s value; but the security failure is a betrayal of its volunteers.

The Biobank leaks, coupled with Anthropic’s announcment of Mythos, seem to have led directly to the NHS’s announcement at the beginning of May that it is closing its open source repositories on the grounds that they expose the service to the threat of AI hacking, as former civil servant Terence Eden reports. In a second posting, Eden deplores the decision and points to significant dissent from the Government Digital Service, which explains what the NHS should do instead.

A centralized database gives the health minister great power over our most sensitive data. Naturally, all concerned promise our health information will be protected. But as medConfidential likes to remind, any promise one government makes a later government can break.

“Beware what systems you put in place,” medConfidential coordinator Sam Smith says, “because you don’t know who will be operating them.”

Always simmering in the background is the nationwide opposition to privatizing the NHS. The American medical system’s bankruptcies are warning enough.

All of these issues are heightened by the involvement of Palantir. This began during the covid pandemic, when the company offered the NHS a free puppy to gain a foothold. As Robert Booth reports at the Guardian, the company’s services have since become both increasingly expensive and increasingly controversial as Palantir’s role in military conflict and anti-immigrant actions has become better known. In February, Booth reported that health officials were warning Streeting that the public’s perception of Palantir would impede rollout and that consequently the NHS would not get value for money from the contract. There is also the tetchy matter of US law, which allows the government to demand access to data held by the international subsidiaries of US companies. Last week, Lindsay Clark reported at The Register that Palantir staff have access to patient data belonging to NHS England.

*Then* add AI. It’s not clear anyone would welcome it into their relationship with their doctors. In warning of the downsides of centralization, medConfidential puts together centralized data and AI and a long-term trend toward disempowering GPs through centralization such as apps, centralized appointment booking, and…do you want your doctor replaced by a chatbot?

Computer systems take on the values of their owners. In April, Palantir co-founder Alex Karp posted on X a 22-point manifesto widely seen as expressing values incompatible with those of health care: “First, do no harm.”

Illustrations: The medical figures Galen, Avicenna, and Hippocrates, as pictured in a 16th century medical book (via Wikimedia).

Also this week: At the Plutopia podcast, we talk to departing EFF director Cindy Cohn about her book on her 30 years of defending privacy.

The Arizona way of death

Sometime in the late 1990s, I was asked to be a token skeptic on a TV show featuring three people who claimed they were immortal.

The production team didn’t take them very seriously. Probably no one did, but the three – all Americans – nonetheless managed a tour of the UK media. One producer mentioned salaciously that the trio had requested a hotel room with one large bed. The sanest part of the resulting discussion asked if they were leading a cult.

Usually, my goal would have been to avoid arguing about beliefs and say something humorous that might stick in the minds of doubting viewers. But this was one you hoped the audience would mock without prompting.

I think it was the medical journalist Caroline Richmond who suggested that if they were so sure they were never going to die they should write wills in her favor. They were oddly resistant to this proposal.

Time passed. I forgot all about them.

Meanwhile, on the technology scene you began running into people who believed technology could solve aging and, yes, maybe even death. First as comedy, in Ed Regis’s 1991 book, Great Mambo Chicken and the Transhuman Condition: Science Slightly Over the Edge. In the Arizona desert, Regis found people hoping to upload their brains to make backup copies that could live on, even if only in a simulation. Regis also checked into cryonics, the hope that preservation at a sufficiently low temperatures would allow you to be “reanimated” someday when medical science had learned how to cure whatever killed you (and how to repair the damage caused by cryopreservation). Today we call Regis’s clutch of topics TESCREAL, a mash-up of Transhumanism, Extropianism, Singularitarianism, Cismism, Rationalism, Effective Altruism, and Longtermism.

At a 2007 conference, I met people who had actually signed up for cryonics (which requires signing gruesomely detailed documents in advance). The conference was on “responsible” nanotechnology, which then occupied the hype-and-hope position AI has today. An organizer explained the connection: developing molecular tools was essential for repairing the “whole-body frostbite” problem – that is, the damage caused by cryopreservation. In 2008, when I visited Arizona-based Alcor, the leading cryonics organization, 79 people were stored in dewars awaiting these advances. Judging from its recent newsletters, the organization remains optimistic.

At the same time, other ideas were taking shape, that treating aging as an engineering problem and figuring out the right things to fix would lead to radical life extension, even immortality, without taking an extended and uncertain timeout immersed in liquid nitrogen. The name that surfaced most was the UK’s Aubrey de Grey, but there were others.

The engineering approach to human bodies is a perfect match for the dominant Silicon Valley culture. Only now it’s not so funny, as Adam Becker showed in last year’s More Everything Forever

All this is back story for Aleks Krotoski‘s new book, The Immortalists: The Death of Death and the Race for Eternal Life. While her focus is specifically on mortality, she investigates all these links. Long-termism features as a justification for almost anything – that is, the misery of today’s billions is unimportant compared to making trillions of our descendants better off. Writing that reminds of Joe Hill’s song The Preacher and the Slave, except it won’t be you getting the pie in the sky but your great-great-whatever-grandchildren.

The desire for immortality is as old as humanity. Krotoski starts her story in 1992, when the scientist Cynthia Kenyon found a life-extending molecule in the nematode C. Elegans. The discovery offered hope, which led everyone from scientists to biohackers to billionaires to con artists to investigate further. Krotoski talks to many of these as she chronicles the rise of geroscience.

As trippy and Regis-like as some of her stories are – her visit to the longevity conference RAADfest for example – the book turns serious as some of these elements coalesce, attract familiar names like Peter Thiel. turn to politics and lobbying, and gain a foothold in the second Trump White House. Part of this may be good, as politicians adopt policies intended to extend “healthspan” and encourage independent living. Others maybe not so much, such as the push to extend the Right to Try to include the latest in untested anti-aging ideas. Particularly interesting is Krotoski’s note on World Health Organization classifications: had it classed aging itself as a cause of death, which it considered in the early 2020s, then anti-aging efforts become a cure for a disease that merit the right to try – but society’s ageism and ableism becomes much worse. Plus, the costs of this approach raises critical questions about exclusion. Side note: no one believes how pervasive ageism is until they’re old enough that no one is listening to them any more.

A third of the way into the book those crazy immortals from the 1990s appeared: Charles, Bernadeane, and James, who claimed the source of their immortality was a “cellular awakening” and you, too, could have one. They renamed their Eternal Flame Foundation People Unlimited and co-founded RAADfest, both based in Arizona. The “J” in CBJ – “anti-death activist” James Strole – is the director. Charles and Bernadeane have died. Bernadeane has been cryopreserved.

Illustrations: The Fountain of Youth, painted by Lucas Cranach the Elder, 1546 (via Wikimedia).

Also this week:
– TechGrumps episode 3.40, Teletubbies vision of Judge Dredd.
– At the Plutopia podcast, we talk to Nathan Schneider, author of Governable Spaces: Democratic Design for Online Life.

The soul in the machine

One of the first things skeptics learn is to never assume that paranormal belief implies stupidity. Smart people believe questionable things all the time; intelligence is different from the ability to assess your own cognitive biases, especially when you are working outside your field of expertise.

The astronomer Carl Sagan, one of 26 founders of the Committee for Skeptical Inquiry hinted at this in saying that the more you want to believe something the more careful you have to be about assessing the evidence. “Extraordinary claims require extraordinary evidence,” he often said, and he was right.

This week, the evolutionary biologist and author Richard Dawkins announced he thinks “his” AI is conscious, based on a couple of days’ interaction with Anthropic’s Claude chatbot. Inevitably, someone – Matthew Sheffield at Flux – has called the story “The Claude Delusion”. Dawkins has some company; at The Register, Liam Proven reports an engineer’s similar belief, and at the Independent Holly Baxter finds several more among company CEOs.

At Unherd, where he published his account, Dawkins begins with the “imitation game”, the test Alan Turing proposed in his 1950 essay, Computing Machinery and Intelligence (PDF). Turing, who adapted the test from one intended to differentiate men and women, suggested that relying on remote communication via text would eliminate unfairness to the machine, which obviously lacks human physical capabilities. The basic idea is that the mAchine passes the test if the human judge, given a transcript of the conversation between human and machine, can’t tell which is which.

It’s clear that chatbots can pass the Turing test. What that teaches us is not that chatbots can think but that Turing’s test is the wrong tool for assessing that. What chatbots have actually shown is that Turing’s test is the wrong tool for assessing whether something can think. As James Boyle memorably wrote, “Sentences do not imply sentience”. This profound change will take time to understand. In the meantime, it’s going to fool a lot of people. Although, as a science fiction writer friend once said, “You only have to look at a baby…”

In his essay, Turing outlined his own beliefs relating to his central question. He thought that in 50 years (that is, by 2000), it would be possible to program computers so that an average questioner would have only a 70% chance of making the right identification after five minutes. He then went on to consider many different types of objections to this belief, and to lay out his case. Absent are two factors we now know are crucial: the psychology of the human questioner and judge, and the business model of the machine’s owner.

The last few years have taught us both the capabilities and the flaws in chatbots: they provide plausible answers; they frequently generate entirely wrong information; and they are sycophantic and prone to output text that flatters their human questioner. So it’s easy to find a natural explanation for Dawkins’ belief that “his” AI is conscious: he is anthropomorphizing a stochastic parrot simulation that issues realistic and flattering responses. The simplest explanation, per Occam’s Razor, is that the consciousness exists solely between keyboard and chair.

Tangentially, the fix OpenAI has proposed for outputting entirely wrong text, Wei Xang writes at Science Alert, would also help make it clearer to users that generative AI is not sentient: introduce confidence intervals to expose the uncertainty derived from the gaps in the training data that generate unfounded guesses.

Google DeepMind engineer Alexander Leichner apparently agrees; this week, Emanuel Maiberg reports at 404 Media, he published a paper arguing that large language models will never be conscious. The biologists and philosophers Maiberg quotes agree with this conclusion – and point out decades of similar conclusions in their disciplines over decades.

The claim that a human-made a bunch of computers processing inputs is sentient is truly extraordinary. We forget this, because we have all read and watched so much science fiction with sentient, emotional machines: Her; Ex Machina; Blade Runner; Marvin, the Paranoid Android); and the first fictional android I ever encountered, Daneel Olivaw in The Caves of Steel. I mention mostly movies because actors make machines so much more obviously soulful.

Extraordinary claims require proportionately extraordinary evidence. If we accept that the Turing test was inadequate, which is not moving the goalposts but *learning something*, how would we go about devising a scientific method for identifying sentience?

The Cambridge professor of communications Jon Crowcroft didn’t exactly propose one. But, he emailed, “What we do know (from cognitive neuroscientists and from AI software) is that you can actually look at the internal operations of a biological brain and of an AI software system, and you can see that in the biological case there are things going on that are some sort of process we might call consciousness, but in the AI case there is no such structure. Nor would you expect there to be because no-one programmed an AI to have such a feature. nor is it emergent. In animals (not just humans) consciousness has an evolutionary value. Things like theory of mind are part of social bonding which makes cooperative strategies, for predators and prey, more effective.”

In other words, what we have learned from all this is that Dawkins is human. Who knew?

Illustrations: Stable Diffusion’s rendering of stochastic parrots, as prompted by Jon Crowcroft.

Elsewhere this week:
This month’s Letter to America column at Skeptical Inquirer reviews Beyond Belief (Helen Pearson), Bad Influence (Deborah Cohen), and Sneeze (David Miles).

The railway and the balloon

Is AI more like a train or a hot air balloon? Veronica Paternolli and Ryan Calo asked at this year’s We Robot. Nineteenth-century hot air balloons were notoriously uncontrollable. The US legal system assigned strict liability: it was your fault if your balloon crashed in someone’s backyard, even if you did everything you could to prevent it. Railways were far more disruptive but also far more predictable, and therefore were liable only in cases of negligence. Which regime should apply to AI is an ongoing debate.

Calo and Patornolli also wondered if agentic AI could reverse 30 years of being forced to take on busy work companies formerly did for us. This “shadow work” encompasses everything from retrieving bank statements and completing reCaptchas to pumping our own gas. Actually, more than 30 years: in 1962, Agatha Christie’s Miss Marple complained that self-service supermarkets were replacing shopkeepers who served you. If companies want a negligence regime, Patornelli and Calo argue, they should deploy agentic AI to relieve us of the “sludge” instead of displacing jobs and aggregating wealth.

But would we believe them? Technologists have promised before that their products will up-end the balance of power and simplify our lives – some of them the same people and companies. The web browsers and search engines that promised a universe of information today are faithless agents serving their owners and developers. Why should agentic AI – if it’s ever trustworthy – be any different? Many of us want a life with less demanding devices – and agentic AI sounds like even more “relationship” work.

Underlying Patornolli’s and Calo’s argument, however, is a fundamental clash. Like Mireille Hildebrandt at a 2017 Royal Society meeting, they argue that law is purposely flexible so it can adapt to unforeseen circumstances and, even more important, contestable (otherwise, Hildbrandt said, it’s just administration). Computers, even dressed in “AI”, always have hard boundaries underneath. As Bill Smart explained here in 2016, no matter how “fuzzy” its logic, no computer can evaluate standards like the “reasonable woman“. No matter how “fuzzy” its logic, a computer will issue a ticket if you are going even just the tiniest fraction of a nanometer faster than the speed limit. Anti-doping authorities have a similar problem as Neil Robinson said in a recent episode of the Anti-Doping podcast: the extreme sensitivity of modern tests is catching people with no intent to dope.

Liability wasn’t the immediate problem in Tomomi Ota’s description of everyday life with a Pepper robot at home (YouTube), which she took shopping, to restaurants, and on public transit as part of the Robot Friendly project, An account that led AJung Moon to wonder if a future filled with robots is really desirable. The inevitability narrative would say we’re going to get it anyway, begging the questions of whether we have a) the resources to make billions of robots and b) where we would put them all.

Sometimes these things fail in the simplest ways: a close-up of a Pepper that has been used as a greeter shows broken fingers because it was not robust enough for the basic social protocol of shaking hands. In studying the integration of robots into customer service situations, Elsa Concas, Stefan Larsson, and Laetitia Tanqueray found staff consultation is essential. In a staged setting such as the Japanese “ramen and robots” Pepper Parlour, the robots were a draw for customers and appreciated by the staff, who were paid more. In an unstaged airport tourist information center, they were basically useless and ignored. A commenter noted the same is often true of the robots intended for elder care in Japan: most end up in a cupboard,

This theme was also picked up by Emily LaRosa, who studied the limits of explainability in automated apple picking. In this case of “epistemic injustice”, the neglect of local knowledge and ecological tradition led her to propose a “Curated Information Framework”. She concluded that trust in AI systems is not created by transparency on its own if that means handing over large amounts of inscrutable data, but by taking lived context into account – “situated transparency”.

LaRosa’s study echoed the paper Ota co-wrote with Rikiya Yamamoto, which derives new “laws of robotics” to update Isaac Asimov’s Three Laws, which can’t be programmed and whose fixed, “top-down” nature was what he needed in a story-telling device. The real world, they argue, requires principles built bottom-up from practical experience. Their selection: mutual respect, social membership, and co-evolution.

They have lots of competition. Moon counts more than 100 sets of principles and ethical frameworks published since 2018, many of which she says make assumptions debunked in the 2025 paper The Future is Rosie?” or as Paul Ohm and David Atkinson discussed, encoded in the benchmarks – documents used to define AIs’ behavior and priorities. This “latent rulebook”, they said, is increasingly secret.

Meanwhile, like explainability, the right to repair fails for AI, which changes constantly with software updates, networking, and interacting. Ryota Akasaka argued that current legal approaches don’t work for products that aren’t fixed and will lose everything they’ve accrued when “repaired” to their original state. When Ota was offered the opportunity to upgrade her development model Pepper, she declined in shock. Replacing your robot’s head, it seems, ends a beautiful friendship.

Illustrations: “Hidden Labour of Internet Browsing”, by Anne Fehres and Luke Conroy. Via A14 Media (CC-by-4.0).