censorship Archives

Fault lines

As widely reported, Meta has settled a lawsuit brought against it by 52 American state and local attorneys general representing most states, the District of Columbia, and several territories The company will, Cecilia Kang and Eli Tan report at the New York Times, pay up to $17.1 billion or, as CNBC has it, $16.7 billion, to be split among the participating states. The higher settlement number includes more than $459 million for claims relating to the Cambridge Analytica scandal. At the Guardian, Aisha Down speculates how the settlement might give weight to the many other claims around the world that remain unaddressed. Even just within the US many other lawsuits relating to child safety are stacked waiting for takeoff.

Few are impressed by the payout; at the Guardian, Chris Stokel-Walker calls it chump change, given that Meta’s profit last year was about £60 billion – and it’s to be paid out over a decade. Seen as more of a win is that the settlement requires change: Meta will impose a two-hour daily usage limit and block use at night for under-18s, enhance age assurance, and provide additional tools for parents and guardians. You can see why plaintiffs are jazzed: settling brings faster results.

Even so, this seems like a win for Meta. It admits no fault. Paying up is cheaper than going on fighting – Stokel-Walker cites estimates provided by Meta’s PR team that legal fees could have reached $1.5 *trillion* (hyperbole, but still). It keeps Meta executives, including CEO Mark Zuckerberg, off the witness stand. And, at Techdirt, Mike Masnick argues that Meta will benefit from the position and ammunition to demand that other platforms such as YouTube and TikTok obey the same rules; it says it will withhold £5 billion of the settlement unless both YouTube and TikTok match. As he writes, the US government could not mandate such changes under the First Amendment.

Both Masnick and the Electronic Frontier Foundation, warn that increased age verification will affect all Meta users in the US. EFF also points out children’s consequential loss of access to information and their speech rights, and gives Meta legal backing to collect even more personal information about its users.

Even without that, as Stokel-Walker notes, the restrictions Meta puts in place for US users are likely to spread as other countries demand equal measures. In the UK, Robert Booth reports at the Guardian that the minister for work and pensions, Pat McFadden, has already called for Meta to extend to British teens changes such as bans preventing under-18s from displaying numbers of likes or reactions, and using cosmetic filters.

At TechCrunch, Sarah Perez spots that the settlement also includes a promise from the attorneys general that they will not sue Meta under the 1998 Children’s Online Privacy Protection Act (COPPA) over its retention and use of children’s data. Meta is now expected to develop, train, and begin testing a model that will detect under-13s using its platforms.

More important, although campaigners say they will continue to push for further changes to enhance child safety, this settlement does nothing to change Meta’s underlying business model, the ultimate source of all the problems. Booth quotes Rachael Kent, who says the changes don’t fundamentally alter the “engagement-driven model” or the algorithms that make it work. As long as the company’s fortunes depend on keeping us all online on its services as long as possible in order to profit from our personal data, change will only be on the surface.

***

Meanwhile, Max Kendix reports at The Times that US Congressman Michael Rulli (R-OH) is sponsoring the GRANITE Act, a bill that would prohibit US courts from recognizing or enforcing foreign fines relating to online speech. That is: it would block Britain’s online safety regulator, Ofcom, from pursuing US companies for failing to comply with the 2023 Online Safety Act. Kendix goes on to say “it is understood” that the bill is supported by senior members of the Trump administration.

The Guaranteeing Rights Against Novel International Tyranny and Extortion bill was first proposed in Wyoming earlier this year. The idea has been pushed for months on X (and presumably elsewhere) by the lawyer Preston Byrne, who is representing 4chan, currently mocking an Ofcom fine of £520,000. Last year, 4chan and Kiwi Farms launched a US lawsuit against Ofcom.

As Madsen Pirie recounts at the Adam Smith Institute, there is precedent in the 2010 SPEECH Act, which protects Americans from foreign libel judgments unless they would have been found liable under US law. At the time, libel tourism was leading plaintiffs to England on the slightest of connections, particularly in a famous case involving Holocaust denialism. Byrne himself we first encountered at the 2015 Tomorrow’s Transactions Forum explaining blockchain.

Byrne’s argument is that US sites with no business presence in Britain are fully protected by the First, Fourth, and Fifth Amendments, and frames the case in terms of Americans’ free speech rights, Ofcom’s position is that sites that are accessible in the UK and meet certain criteria must perform age verification. Seems like the simpler, cheaper response would be to do the same as much less controversial sites, like Imgur: block UK access.

The splintering of the Internet continues.

Illustrations: The Imgur home page, as seen from the UK in 2025.

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.

Eating the web

“Traffic to my blog has plummeted,” a friend said recently. Over decades, he’s built a thriving community, and his core users persist. But Google was crucial for bringing in new readers – and its introduction of AI and changes to its algorithm have punished small sites.

This week, Sean Hollister reports at The Verge that Google is using its AI to replace the headlines on news stories. For Google, it is, as Charles Arthur comments at The Overspill, a “small” and “narrow” experiment – until it becomes a feature. For The Verge, however, the impact is noticeable: the headlines it crafts to market its journalists’ work are being replaced with boring titles that do not accurately convey the articles’ content.

Then, the week before last, Joe Toscano reported at Forbes that Google has patented a system that uses AI to rewrite website landing pages to produce customized versions for its users. Toscano links this to an earlier announcement in which Google announced a protocol to make websites’ structure more readable to AI agents. Toscano suggests that taken together the two elements allow Google to break websites apart into their component parts for reassembly by AI agents into whatever version they identify as best for the user they represent.

In the early 1990s, someone I met directed me to read the work of Robert McChesney, whose books recount the cooption and commercialization of radio and television, also originally conceived as democratic, educational media. Helping to prevent a similar outcome for the Internet is a lot of what net.wars has always been about. Now, Google, which would not exist without the open web, wants to eat the whole thing.

***

On Tuesday, a jury in a New Mexico court found Meta guilty of misleading consumers about the safety of its platforms and enabling harm including child sexual exploitation, as Katie McQue reports at the Guardian. The jury has ordered the company to pay $375 million in civil penalties. Meta will appeal. Snapchat and TikTok, which were also accused, settled before the trial began.

The New Mexico attorney general’s office says it intends to pursue changes to platform design including age verification and “protecting minors from encrypted communications that shield bad actors”.

On Wednesday, a jury in Los Angeles found YouTube guilty of deliberately designing an addictive product. As Dara Kerr reports at the Guardian, the case was brought by a 20-year-old woman who claimed her addiction to Instagram and YouTube began at age six, damaging her relationships with her family and in school and causing her to become depressed and engage in self-harm. The jury awarded her $6 million, split between Meta (70%) and YouTube (30%). Both companies say they will appeal.

They will have to, because, as Kerr reported in January, there are more of these trials to come, and even to trillion-dollar companies thousands of fines can add up to real money. In a consolidated case, in California state and federal courts thousands of families accuse social media companies of harming children. Reuters reports that more trials are scheduled: a school district in Breathitt County, Kentucky in federal court against Meta, ByteDance, Snapchat, and Google, and one in state court in California in July against Instagram, YouTube, TikTok, and Snapchat.

In January, the Tech Oversight Project reported newly unsealed documents contained the “smoking gun” evidence – that is, internal email discussions – that the four companies deliberately designed their products to be addictive and failed to provide effective warnings about social media use. Certainly, the leaked documents make it sou9nd like a plan. Tech Oversight quotes one: “Onboarding kids into Google’s Ecosystem leads to brand trust and loyalty over their lifetime.” It’s hard not to see the commonality with Joe Camel and so many other marketing strategies.

Key to these cases is Section 230 – the clause in the Communications Decency Act that shields online services from liability for the material their users post and allows them to moderate content in good faith. The plaintiffs argued – successfully in New Mexico – that the law does not shield the platforms from liability for their design decisions. The social media companies naturally tried to argue that it does.

At his blog, law professor Eric Goldman discusses the broader impact of these bellwether cases. As he says, whatever changes the social media companies feel forced to make by the potential liability of myriad jury trials and new laws may help some victims but almost certainly hurt other groups who were not represented at the trial. Similarly, at Techdirt Mike Masnick warns that features like autoscrolling and algorithmic recommendations are not inherently harmful; it’s the content they relentlessly serve that is really the issue; cue the First Amendment. And few who are not technology giants can afford to face jury trials and fines. Are we talking a regime under which every design decision has to go through lawyers?

In a posting summarizing the history of S230, Goldman predicts that age verification laws will reshape the Internet of 2031 or 2036 beyond recognition, killing most of what we love now. So much doom, so little time.

Illustrations: The volcano of Stromboli, on which JRR Tolkien based Mount Doom in The Lord of the Rings (by Steven J. Dengler at Wikimedia.

The censorship-industrial complex

In a sign of the times, the Academy of Motion Picture Arts and Sciences has announced that in 2029 the annual Oscars ceremony will move from ABC to YouTube, where it will be viewable worldwide for free. At Variety, Clayton Davis speculates how advertising will work – perhaps mid-roll? The obvious answer is to place the ads between the list of nominees and opening the envelope to announce the winner. Cliff-hanger!

The move is notable. Ratings for the awards show have been declining for decades. In 1960, 45.8 million people in the US watched the Oscars – live, before home video recording. In 1998, the peak, 55.2 million, after VCRs, but before YouTube. In 2024: 19.5 million. This year, the Oscars drew under 18.1 million viewers.

On top of that, broadcast TV itself is in decline. One of the biggest audiences ever gathered for a single episode of a scripted show was in 1983: 100 million, for the series finale of M*A*S*H. In 2004, the Friends finale drew 52.5 million. In 2019, the Big Bang Theory finale drew just 17.9 million. YouTube has more than 2.7 billion active users a month. Whatever ABC was paying for the Oscars, reach may matter more than money, especially in an industry that is also threatened by shrinking theater audiences. In the UK, YouTube is second most-watched TV service ($), after only the BBC.

The move suggests that the US audience itself may also not be as uniquely important as it was historically. The Academy’s move fits into many other similar trends.

***

During this week’s San Francisco power outage, an apparently unexpected consequence was that non-functioning traffic lights paralyzed many of the city’s driverless Waymo taxis. In its blog posting, the company says, “While the Waymo Driver is designed to handle dark traffic signals as four-way stops, it may occasionally request a confirmation check to ensure it makes the safest choice. While we successfully traversed more than 7,000 dark signals on Saturday, the outage created a concentrated spike in these requests. This created a backlog that, in some cases, led to response delays contributing to congestion on already-overwhelmed streets.”

Friends in San Francisco note that the California Driver’s Handbook (under “Traffic Control”) is specific about what to do in such situations: treat the intersection as if it had all-way stop signs. It’s a great example of trusting human social cooperation.

Robocars are, of course, not in on this game. In an uncertain situation they can’t read us. So the volume of requests overwhelmed the remote human controllers and the cars froze, blocking intersections and even sidewalks. Waymo suspended the service temporarily, and says it is updating the cars’ software to make them act “more decisively” in such situations in future.

Of course, all these companies want to do away with the human safety drivers and remote controllers as they improve cars’ programming to incorporate more edge cases. I suspect, however, that we’ll never really reach the point where humans aren’t needed; there will always be new unforeseen issues. Driving a car is a technical challenge. Sharing the roads with others is a social effort requiring the kind of fuzzy flexibility computers are bad at. Getting rid of the humans will mean deciding what level of dysfunction we’re willing to accept from the cars.

Self-driving taxis are coming to London in 2026, and I’m struggling to imagine it. It’s a vastly more complex city to navigate than San Francisco, and has many narrow, twisty little streets to flummox programmers used to newer urban grids.

***

The US State Department has announced sanctions barring five people and potentially their families from obtaining visas to enter or stay in the US, labeling them radical activists and weaponized NGOs. They are: Imran Ahmed, an ex-Labour advisor and founder and CEO of the Centre for Countering Digital Hate; Clare Melford, founder of the Global Disinformation Index; Thierry Breton, a former member of the European Commission, whom under secretary of state for public diplomacy Sarah B. Rogers, called “a mastermind” of the Digital Services Act; and Josephine Ballon and Anna-Lena von Hodenberg, managing directors of the independent German organization HateAid, which supports people affected by digital violence. Ahmed, who lived in Washington, DC, has filed suit to block his deportation; a judge has issued a temporary restraining order.

It’s an odd collection as a “censorship-industrial complex”. Breton is no longer in a position to make laws calling US Big Tech to account; his inclusion is presumably a warning shot to anyone seeking to promote further regulation of this type. GDI’s site’s last “news” posting was in 2022. HateAid has helped a client file suit against Google in August 2025, and sued X in July for failing to remove criminal antisemitic content. The Center for Countering Digital Hate has also been in court to oppose antisemitic content on X and Instagram; in 2024 Elon Musk called it a ‘criminal organization’. There was more logic to”the three people in hell” taught to an Irish friend as a child (Cromwell, Queen Elizabeth I, and Martin Luther).

Whatever the Trump administration’s intention, the result is likely to simply add more fuel to initiatives to lessen European dependence on US technology.

Illustrations: Christmas tree in front of the US Capitol in 2020 (via Wikimedia).

The safe place

For a long time, fear that technical decisions – new domain names ($), cooption of open standards or software, laws mandating data localization – would splinter the Internet. “Balkanize” was heard a lot.

A panel at the UK Internet Governance Forum a couple of weeks ago focused on this exact topic, and was mostly self-congratulatory. Which is when it occurred to me that the Internet may not *be* fragmented, but it *feels* fragmented. Almost every day I encounter some site I can’t reach: email goes into someone’s spam folder, the site or its content is off-limits because it’s been geofenced to conform with copyright or data protection laws, or the site mysteriously doesn’t load, with no explanation. The most likely explanation for the latter is censorship built into the Internet feed by the ISP or the establishment whose connection I’m using, but they don’t actually *say* that.

The ongoing attrition at Twitter is exacerbating this feeling, as the users I’ve followed for years continue to migrate elsewhere. At the moment, it takes accounts on several other services to keep track of everyone: definite fragmentation.

Here in the UK, this sense of fragmentation may be about to get a lot worse, as the long-heralded Online Safety bill – written and expanded until it’s become a “Frankenstein bill”, as Mark Scott and Annabelle Dickson report at Politico – hurtles toward passage. This week saw fruitless debates on amendments in the House of Lords, and it will presumably be back in the Commons shortly thereafter, where it could be passed into law by this fall.

A number of companies have warned that the bill, particularly if it passes with its provisions undermining end-to-end encryption intact, will drive them out of the country. I’m not sure British politicians are taking them seriously; so often such threats are idle. But in this case, I think they’re real, not least because post-Brexit Britain carries so much less global and commercial weight, a reality some politicians are in denial about. WhatsApp, Signal, and Apple have all said openly that they will not compromise the privacy of their masses of users elsewhere to suit the UK. Wikipedia has warned that including it in the requirement to age-verify its users will force it to withdraw rather than violate its principles about collecting as little information about users as possible. The irony is that the UK government itself runs on WhatsApp.

Wikipedia, Ian McRae, the director of market intelligence for prospective online safety regulator Ofcom, showed in a presentation at UKIGF, would be just one of the estimated 150,000 sites within the scope of the bill. Ofcom is ramping up to deal with the workload, an effort the agency expects to cost £169 million between now and 2025.

In a legal opinion commissioned by the Open Rights Group, barristers at Matrix Chambers find that clause 9(2) of the bill is unlawful. This, as Thomas Macaulay explains at The Next Web, is the clause that requires platforms to proactively remove illegal or “harmful” user-generated content. In fact: prior restraint. As ORG goes on to say, there is no requirement to tell users why their content has been blocked.

Until now, the impact of most badly-formulated British legislative proposals has been sort of abstract. Data retention, for example: you know that pervasive mass surveillance is a bad thing, but most of us don’t really expect to feel the impact personally. This is different. Some of my non-UK friends will only use Signal to communicate, and I doubt a day goes by that I don’t look something up on Wikipedia. I could use a VPN for that, but if the only way to use Signal is to have a non-UK phone? I can feel those losses already.

And if people think they dislike those ubiquitous cookie banners and consent clickthroughs, wait until they have to age-verify all over the place. Worst case: this bill will be an act of self-harm that one day will be as inexplicable to future generations as Brexit.

The UK is not the only one pursuing this path. Age verification in particular is catching on. The US states of Virginia, Mississippi, Louisiana, Arkansas, Texas, Montana, and Utah have all passed legislation requiring it; Pornhub now blocks users in Mississippi and Virginia. The likelihood is that many more countries will try to copy some or all of its provisions, just as Australia’s law requiring the big social media platforms to negotiate with news publishers is spawning copies in Canada and California.

This is where the real threat of the “splinternet” lies. Think of requiring 150,000 websites to implement age verification and proactively police content. Many of those sites, as the law firm Mischon de Reya writes may not even be based in the UK.

This means that any site located outside the UK – and perhaps even some that are based here – will be asking, “Is it worth it?” For a lot of them, it won’t be. Which means that however much the Internet retains its integrity, the British user experience will be the Internet as a sea of holes.

Illustrations: Drunk parrot in a Putney garden (by Simon Bisson; used by permission).

Wendy M. Grossman is the 2013 winner of the Enigma Award. Her Web site has an extensive archive of her books, articles, and music, and an archive of earlier columns in this series. Follow on Mastodon.