We are developing the social individualist meta-context for the future. From the very serious to the extremely frivolous... lets see what is on the mind of the Samizdata people.
Samizdata, derived from Samizdat /n. - a system of clandestine publication of banned literature in the USSR [Russ.,= self-publishing house]
|
The BBC says “According to Ofcom, platforms must not host, share or permit content encouraging use of VPNs to get around age checks.”
I encourage the use a VPN to get around all state abridgement of people’s right to access the internet, including age checks. Say no to police state Britain, not to mention a VPN enhances your security online.
The free and open internet has now ceased to exist in the UK. Since Friday, anyone in Britain logging on to social media will have been presented with a censored, restricted version – a ‘safe’ internet, to borrow the UK government’s language. Vast swathes of even anodyne posts are now blocked for the overwhelming majority of users.
The Online Safety Act was passed by the last Conservative government and backed enthusiastically by Labour. Both parties insisted it is necessary to protect children. Supposedly, its aim is to shield them from pornography, violence, terrorist material and content promoting self-harm. Age-verification checks, we were assured, would ensure that children would not be exposed to inappropriate content, but adults could continue using the internet as they please. Yet as we have seen over the past few days, on many major tech platforms, UK-based adults are being treated as children by default, with supposedly ‘sensitive’ content filtered from everyone’s view.
– Fraser Myers
Police state Britain needs nothing less than a revolution.
When Israel struck Iranian targets on June 13th, something strange happened four days later—thousands of Twitter accounts tweeting about Scottish independence just… stopped.
That silence? It wasn’t just suspicious. It was evidence.
In this video, I walk you through how a digital blackout in Iran exposed one of the largest Iranian disinformation operations targeting the West. Working with Cyabra, we tracked over 1,300 fake accounts—AI-generated personas pushing division in the UK, attacking the BBC, and praising Iran. And when the lights went out in Tehran, those bots vanished too.
Then they came back… parroting entirely new propaganda.
This is a masterclass in how modern influence warfare works—how state-sponsored actors weaponize your feed and how even silence can be a signal.
– Ryan McBeth
In December 2023, I asked “Non-sarcastically, why am I so sure that this image is generated by AI?” and listed the reasons why I thought that a picture purporting to show gleeful Israeli soldiers in Gaza was a fake.
In July 2025, I must modify my question. Why am I mostly sure that this image, also purporting to show events in Gaza, is generated by AI?

I saw the picture in a Telegraph story written by Melanie Swan and called “More than 90 dead in UN aid truck massacre in Gaza”. The caption says, “Injured Palestinians are taken to hospital after over 90 were killed waiting for humanitarian aid Credit: Ali Jadallah/Anadolu via Getty”.
Here are some of the reasons why I think the picture is AI-generated:
I always start by counting their fingers and toes. The left foot of the guy holding his knee appears to have six of the latter.
The little toe of the left foot of the bare-legged boy sitting in the centre looks wrong; too wide, no toenail – just a wedge of flesh.
Staying with the boy, his legs seem malformed – the distance from knee to ankle too long, the thigh too short and too narrow.
His right arm is too short and floppy, like the vestigial arm of a Tyrannosaurus Rex.
The little finger of the hand of the guy with the beard seated on the right of the picture does not join up to the hand correctly – either that or he was unfortunate enough to be born with his left hand where his right hand ought to be.
The writing on that white bag on the shelf is blurred in a way characteristic of A.I.
Moving back to the left of the picture, where is the long, thin arm pointing diagonally downwards coming from? In what position would a person be lying or standing in order to have their arm come out at that angle?
Compare the thin arm to the arms of the other people in the picture. It looks too long and thin to be true; an adult’s arm would be thicker, a child’s arm would be shorter. And, though I strain to see it, the hand looks almost as if it has two thumbs.
If you look at the picture under high magnification, it looks almost like someone has drawn around the figures with a Sharpie. These black outlines are particularly noticeable with the long-legged boy and the man clutching his knee.
This one is more speculative, but do the interiors of Israeli or Palestinian ambulances actually look like that? The (oddly sparse) contents of the shelves suggest a medical purpose, but the shelves themselves look like they come from someone’s kitchen.
Taken separately, all of the above points could be explained away. Lenses distort. Human bodies vary. Hunger makes people thin. Perhaps I will end up deleting this post in shame at having questioned the suffering of real human beings. Perhaps, but, having been able to find at least eight oddities, I think that Getty Images would be justified in putting a few pointed questions to Ali Jadallah.
However, I was right to say in 2023 that “this image is a great deal more realistic than those of only a few months ago. My spidey-sense for fake pictures will not last much longer”. It is even more true now.
Added 22/07/2025: Reading the comments to the Telegraph article, a lot of the commenters are saying, like me, that one of the photographs the Telegraph has used to illustrate it is fake. Only they are talking about a different picture. This one:
It shows a boy running away while a cloud of smoke rises from the buildings behind him.
With the picture of the men in the vehicle that I talked about above, my suspicions were raised the instant I saw it. The hyper-defined outlines and sharp colours gave a sort of slick, sweaty appearance to the flesh of the people depicted that I have often seen in A.I. art and noticed on that picture even before I started counting their digits. There is nothing like that in this second picture. The strange things about it suggest Photoshop rather than DALL-E or Midjourney. The border of the smoke cloud is at a suspiciously neat 45 degree angle. There is also something suspicious about the way the buildings to the left of the boy merge into the smoke. But the main problem is the running boy himself. The photographer appears to have caught him in mid-air – fine, that can happen when taking a photo of a person leaping or running, and catching that moment is usually considered the mark of a successful, dramatic picture – but he is too high off the ground to be plausible. And he has no shadow.
Or does he? There are two darker almost-horizontal lines or one slightly bent line below and to the left of him that could be his shadow. And before anyone brings up the similar horizontal lines to the right of him, those could be the shadow of a tree or pole just outside the picture. He is still suspended at an unlikely height, though. All in all, I am less convinced of the fakeness of this picture than of the other one – and the whole point of this post was that it is getting harder and harder to tell.
We are entering an age in which decisive authentication of a photograph will no longer be possible. The question will be whether one trusts the source. I do not trust anything coming out of Gaza.
Sadiq Khan, the Labour mayor of London, and Anne Hidalgo, the Parti Socialiste mayor of Paris, have written a joint article for the Guardian called “In London and Paris, we’ve experienced vicious backlash to climate action. But we’re not backing down”. They write,
“We welcome efforts such as the EU’s Digital Services Act, which requires online platforms to counter the spread of illegal content, including disinformation, and lays the groundwork for holding platforms accountable. But much more is needed. For example, the UK’s Online Safety Act could be strengthened by explicitly recognising climate disinformation as a form of harmful content.”
It is remarkable how people who would be ashamed to support a law to “counter the spread of illegal speech” happily praise a law that “counters the spread of illegal content.” The magic of words: just re-label “speech” as “content” – as being inside something – and it can now truly be contained, as in “restrained or controlled”.
The same people regularly proclaim that Europe is a place which has banished censorship of the press. That is almost true, although both the EU and the UK governments are working to restore their old powers. In the meantime they are willing enough to temporarily refrain from censorship of ideas spread by old technology if it gives them cover for censoring ideas spread by new technology.
Do not go along with their word games. The term “Freedom of the press” is not restricted to words conveyed to the public by means of a a mechanical device for applying pressure to an inked surface resting upon a print medium. Nor does “freedom of speech” only refer to words that come out of mouths by the action of tongue and lips. In the words of a source they claim to respect, Article 19 of the Universal Declaration of Human Rights says,
Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers.
Emphasis added.
And while I’m emphasising things, let me also emphasise this: the minute I learn that an idea is being censored, I give that idea more credence.
No, that does not mean that I automatically believe any censored idea entirely. (How could I? A million contradictory falsehoods are censored alongside the truth. The problem is that the act of censorship destroys our ability to tell which of them is the truth.) It means that I strain to hear what is being said behind the gag. It means that I start to wonder how real the claimed consensus is, if those who depart from it are silenced. It means that I start to wonder why the proponents of the “accepted” view feel the need to protect it from counter-arguments.
I said in 2012 that my belief in Catastrophic Anthropogenic Global Warming (CAGW) was two and a half letters to the left compared to most commenters on this blog. Damn, it should have been one and a half letters to the right. Oh well, you knew what I meant. With regard to CAGW or whatever they are calling it now, whether I phrase it as my belief moving to the left or my disbelief moving to the right, the surest way to make that movement happen is to pass a law defining “climate disinformation as a form of harmful content”. Then I will know that the so-called scientific consensus on climate change is no such thing. If certain hypotheses cannot be discussed, not only is there no scientific consensus, there is no science.
-*-
A related post, but focussing on self-censorship rather than the government censorship that Mayors Khan and Hidalgo favour: “Bubbles, lies and buttered toast.” Any form of censorship is fatal to science.

Nice one Eyal! This is great work and genuinely hilarious 😀
Asmongold is a Twitch streamer whose output is also edited and put on YouTube.
A funny example (specifically the timestamp at 14:20).
He is making fun of some protestors. At first glance it is inane. But there is more here.
He makes several points. Free speech and peaceful protest are important. As soon as people are setting fire to things, it is no longer a protest but a riot. Rioters should be dealt with swiftly and severely to discourage others. Blocking the highway or taking over buildings is infringing on others’ rights.
Here he is covering conflict between India and Pakistan.
This is how The Kids are getting The News These Days. Streamers are surfacing, and commenting on, both mainstream and social media content.
This is no bad thing. Mainstream media getting to set the narrative has proven unhealthy. Blogs had their day. Video is now where it’s at. Streamers and influencers are filtering things.
This could be good or bad. It depends on the streamer. Asmongold is thoughtful, non-partisan, exercises critical thinking, caveats and bounds his opinions, avoids giving opinions where he lacks knowledge (such as specifics of politics between India and Pakistan in the above example), avoids (when he is being serious) sweeping generalisations, has views mostly compatible with maximising freedom and in general seems pretty smart.
That he is one of the most successful and influential at doing this, is more successful than others in a similar line of work who might charitably be considered dangerous idiots, gives hope that the natural filter of the algorithms can do good.
As most of you will know, I covered the sentencing of the Southport Killer live on Twitter/X as event unfolded in the courtroom on 23 January 2025.
During the hearing, I created a timeline recounting what happened on the day of the attack, minute-by-minute, so that the public could see the full horror of this attack, and what had been kept out of the media.
This was followed by indirect criticism from Merseyside Police who claimed the families had asked for the details of the case not to be published. This had been a lie, told for the convenience of the Police who did not want a riot to breakout as a result of their lies and inaction.
– Charlie Bentley-Astor
Read the whole thing.
Keep comments relevant.
Why? How many real-life, off-screen cases of femicide has Tate actually been provably linked with? Not as many as a casual newspaper reader may be led to presume. Andrew didn’t bomb all those little girls to death at the Manchester Arena a few years back, did he? Mere days after Adolescence went up on Netflix, the UK’s counter-terrorism tsar, Robin Simcox, released a report into 100 convicted UK-based terrorists arrested between 2004 and 2021, analysing their “mindset material”, like social media activity. This found that, of the 100 studied, 85 could be classed as Islamists, 14 as ‘far-Right’ (whatever that even is now) and… one as being an incel. Appropriately enough, really, for such a committed breed of professional loners.
– Steven Tucker (£)
The political scientist Timur Kuran coined the term “preference falsification” in 1987. Earlier today he sent this tweet:
Musk’s purchase of Twitter was a political game changer. Also important was his decision to hide people’s “likes” from other users. This diminished preference falsification on X. It also boosted the apparent popularity, and thus the circulation, of un- or anti-woke posts.
Tony Blair greatly increased the ease of postal voting in UK elections by means of the Representation of the People Act 2000. That Wikipedia article says the Act made only “minor amendments”. They were not minor in their effects and nor were they intended to be. Whoever edited the Wikipedia article on Absentee voting in the United Kingdom got it right:
After the introduction of on-demand postal voting in the UK, there has been a massive uptake in postal voting. Whilst in 2001 1.8 million postal ballots were distributed to voters, this has increased to more than 8 million postal ballots by the UK 2017 general election and represented one in every five ballots cast in 2019 United Kingdom general election.
Labour did this because they thought it would help them win elections, of course. Did it? Perhaps not. While it did increase turnout, which historically has usually helped Labour candidates, the increase in turnout was particularly strong among pensioners, who tend to have mobility problems that make it harder for them to get to the polling station in person. Pensioners skew Conservative. The change also had other effects, of which more below.
I can certainly see a reason for some mechanism to be available to let people arrange to vote by post (or vote by mail as the Americans call it) when circumstances make them unable to vote in person. But absentee voting unquestionably degrades the secret ballot. This brings us back to the issue of preference falsification. As the same Wikipedia article says,
In the United Kingdom a 2016 government inquiry found that postal voting “was considered by some to be the UK’s main electoral vulnerability and to provide the ‘best’ opportunity for electoral fraud… Evidence was presented of pressure being put on vulnerable members of some ethnic minority communities, particularly women and young people, to vote according to the will of the elders… the possibilities of undue influence, theft of postal votes and tampering with them after completion were all still risks.” The government responded by saying it would consider the recommendations on postal voting.
Presumably the government (by then a Conservative one) did consider the recommendations. It evidently decided it wanted more postal voting anyway. Probably that was to get the pensioner vote.
However something changed in the 2024 election that I speculate might lead Labour to fall out of love with postal voting. Of course Labour won that election with a massive majority – but there were some nasty surprises for individual Labour MPs, many of them quite prominent.
Wes Streeting, the Secretary of State for Health, had a majority of 5,218 in the 2019 election. His majority in the 2024 election was 528. The person who came near to unseating him was a Muslim Independent who campaigned on the issue of Gaza.
Shabana Mahmood, the Secretary of State for Justice, had a majority of 28,582 in the 2019 election. Her majority in the 2024 election was 3,421. The person who came near to unseating her was a Muslim Independent who campaigned on the issue of Gaza.
Jess Philips had a majority of 10,659 in the 2019 election. Her majority in the 2024 election was 693. The person who came near to unseating her is a Muslim member of George Galloway’s Workers Party who campaigned on the issue of Gaza.
Jonathan Ashworth had a majority of 22,675 in 2019. His constituency was considered a safe seat for Labour, but he lost it in 2024 to a Muslim Independent who campaigned on the issue of Gaza.
There are several other similar examples.
Labour knows full well that its current majority is a mile high but an inch thick, as the saying goes. If Reform eats the Tories, or vice versa, I think that Labour will look with fresh eyes at the issue highlighted in that 2016 report:
Evidence was presented of pressure being put on vulnerable members of some ethnic minority communities, particularly women and young people, to vote according to the will of the elders.
Sometimes the Guardian shows flashes of its old persona as a guardian of liberty. Publishing this article by Apostolis Fotiadis was one example:
The EU wants to scan every message sent in Europe. Will that really make us safer?
In my 20 years of being a reporter, I have rarely come across anything that feels so important – and yet so widely unnoticed. I’ve been following the attempt to create a Europe-wide apparatus that could lead to mass surveillance. The idea is for every digital platform – from Facebook to Signal, Snapchat and WhatsApp, to cloud and online gaming websites – to scan users’ communications.
This involves the use of technology that will essentially render the idea of encryption meaningless. The stated reason is to detect and report the sharing of child sexual abuse material (CSAM) on digital platforms and in their users’ private chats. But the implications for our privacy and security are staggering.
Since 2022, EU policymakers have attempted to push the legislation, called the regulation to prevent and combat child sexual abuse (better known as the CSAM regulation proposal), through. Similar attempts to introduce the tech in Britain via the online safety bill were abandoned at the 11th hour, with the UK government admitting it is not possible to scan users’ messages in this way without compromising their privacy.
Cybersecurity experts have already made their opinions clear. Rolling out the technology will introduce flaws that could undermine digital security. Researchers based at Imperial College London have shown systems that scan images en masse could be quietly tweaked to perform facial recognition on user devices without the user’s knowledge. They have warned there are probably more vulnerabilities in such technologies that have yet to be identified.
The title of this post referred to this story: “Britain’s biggest choir ditches Every Breath You Take over ‘abusive’ lyrics”
The song, which was written by Sting and released in 1983, is considered by some to be a stalkers’ anthem.
Sting has admitted that the words – “Every breath you take/ And every move you make/ Every bond you break/ Every step you take/ I’ll be watching you” – have “sinister” overtones.
I have been predicting this day would come for decades. It is still chilling to see it arrive.
Both today’s Sunday Times and the Mail on Sunday carry the story of an ordinary woman whose life was nearly ruined by an AI-edited version of some doorbell footage that falsely showed her uttering racist abuse. The Mail’s story is here. It has the original video without a paywall, but I had started writing this post using the Sunday Times version before I was made aware by commenter JuliaM that the Mail had the same story, so in what follows I will mostly quote the Sunday Times story, ‘I doorknocked for Labour then racist deepfake ruined my life’. An archived version can be found here.
It started harmlessly enough. A PE teacher called Cheryl Bennett said that she would help deliver leaflets for her colleague, Quasim Mughal, who was standing as a Labour candidate in the local elections in May last year.
For that display of friendship, she has paid a heavy penalty. What happened that morning — or, rather, did not happen — has changed her life forever. For a time, it cost Bennett her reputation and her career. She was at risk of a criminal conviction too, and police visited her home to arrest her.
As she approached the door of a household in nearby Dudley, she was accompanied by two people: Mughal, the candidate who is of south Asian heritage, and her previous head teacher, who is not. At first, the owner did not answer.
By the time the door was opened, both colleagues had moved on to the next property, leaving Bennett to ask the person whether they intended to vote. Unbeknown to her, a CCTV camera perched above the door was filming.
Within days, a short segment of the footage had been leaked, edited to remove Mughal, and given subtitles. The resulting video falsely depicted Bennett launching into a racist tirade against the homeowner, with subtitles declaring: “F***ing p*kis. P*kis,” as she walked away from the front door.
Nobody has been able to establish who maliciously doctored the footage, but it was given to Akhmed Yakoob, a Lamborghini-owning criminal solicitor, nicknamed the “TikTok lawyer”, who was an independent pro-Gaza candidate for West Midlands mayor and had close links with George Galloway’s Workers Party of Britain.
Yakoob posted a narrated version of the fake video on TikTok. He also posted Bennett’s name and place of work.
The video caused a sensation. Within days, it had received 2.1 million views across TikTok, Facebook and X, and prompted hundreds of people, including dozens of parents at her school, which has a large British-Pakistani community, to demand she be sacked. Yakoob and his followers cited Bennett as an example of Labour and Sir Keir Starmer’s lack of interest in Muslim and minority ethnic voters in the wake of the Israel-Gaza war. She was forced into hiding.
Yakoob has since paid substantial damages for his publication of the video.
For a time, however, it looked as though vindication might never come. Within a short time of Yakoob’s TikTok post at 7.30pm, her phone started to vibrate while she was at a friend’s house.
“My phone just started going off like I’d just stepped out of Love Island or I’d just become famous. It was going absolutely berserk on the table. So I picked it up thinking: ‘Family, is there something going on?’ So I looked at my phone and I had loads of work emails going through.”
Most of them contained abuse. Some were written by children at her own school. “Appalling,” one pupil said. “Being racist is harmful because it disregards the inherent worth and dignity of individuals solely based on their race.” Another wrote: “I didn’t expect a teacher of your standard to be discriminative of races.” Bennett, confused, protested that she had said no such thing, but the messages kept on coming through Facebook, Instagram and LinkedIn. “Stop lying.” “Ur not getting away with this.” “Racist little bitch.”
Then came the formal complaints, as well-meaning parents wrote to the head teacher demanding an investigation and threatening to contact the board of governors. The secondary school received 800 complaints in a short time, some from parents at her school, others from her previous school.
Within hours, the head teacher had told Bennett not to return to work for her own safety. She was not safe at home either, where she lived alone. Strangers arrived at the homes of her parents and her grandmother demanding information as to her whereabouts. Even her car number plate was circulating online.
She stayed at a friend’s home that night. At about 2.30am, West Midlands police went to her home to arrest her, putting a postcard through her door asking her to call them.
The Mail’s version of the story makes it clearer that Ms Bennett having fled to a friend’s house was the reason that she was not present when the police arrived at her home to arrest her at 2.30am. Even if she had been guilty, I do not see why the police thought it was necessary to turn up at that hour to arrest a woman for a non-violent crime.
The Sunday Times account continues:
“I was just constantly in survival mode. I was just trying to get through every single day. And it’s only because I’ve been raised by a very strong family, by very strong women, in terms of you keep fighting and pushing through. Because there was days where I just thought: ‘Would it be easier if I was to just end my life?’ Just because I felt like my career would never be same.”
Before long, police discovered the video was a hoax. They obtained the original doorbell footage, which specialist officers could see bore no resemblance to the subtitles in the video. On May 8, a spokesman for the force said they had found “no evidence of any racist slurs or language used”.
Lucky for her the original footage was still available. How long do they keep it on file? Round here we tend to assume surveillance is bad in itself, but we may soon end up being grateful for it more often than not.
|
Who Are We? The Samizdata people are a bunch of sinister and heavily armed globalist illuminati who seek to infect the entire world with the values of personal liberty and several property. Amongst our many crimes is a sense of humour and the intermittent use of British spelling.
We are also a varied group made up of social individualists, classical liberals, whigs, libertarians, extropians, futurists, ‘Porcupines’, Karl Popper fetishists, recovering neo-conservatives, crazed Ayn Rand worshipers, over-caffeinated Virginia Postrel devotees, witty Frédéric Bastiat wannabes, cypherpunks, minarchists, kritarchists and wild-eyed anarcho-capitalists from Britain, North America, Australia and Europe.
|