The Blog of Extraordinary Anti-Clankers
The problem with AI restoration
Tues 31st March 2026
“Restoration” vs. “Conservation”
For us anti-clankers this article provides a powerful, philosophical argument that goes beyond typical complaints about inaccuracy. The author frames the issue as one of ethics, respect, and the fundamental difference between restoration and conservation.
The very concept of “restoring” old photos with AI is fundamentally flawed. The author draws a parallel to the philosophy of art conservationists at a prestigious institute in Florence, Italy. These experts reject the word “restoration” because it implies actively adding or subtracting from the original work. They prefer “conservation,” which means protecting what exists, preventing further damage, and honoring the piece’s history without altering it.
AI photo “restoration” isn’t a careful process; it’s a brute-force act of invention, not preservation.
Why AI Makes the Problem Worse
The article argues that if a human artist cannot ethically “restore” a photo without imposing their own biases, then an AI certainly cannot. An AI is a “blunt instrument” that has no understanding of history, context, or the subject’s identity. It simply fills in missing details based on its training data, which is a form of digital fabrication.
This leads to two critical failures:
- Grotesque Inaccuracy: When the AI fails, the results are bizarre and unnatural, turning family photos into “nightmare fuel.” This is the most obvious problem.
- Identity Theft (The Subtler, More Sinister Problem): Even when the AI’s results look plausible, they are often fabrications. The article provides a powerful example where an AI “restored” a vintage portrait, changing the subject’s jacket color and facial features to the point where he is no longer the same person. In a group photo, one man’s face was altered so completely that his “identity was stolen.” The AI didn’t restore the person; it created a new one based on a vague template.
The Process, Not the Technology, Is Broken
The company behind the AI tool, ON1, responded to criticism by claiming the examples were poor and the tool was meant to be part of a larger workflow. The author dismisses this as irrelevant, arguing that the fundamental problem is the process of AI generation itself, not the quality of a specific tool.
This is a crucial point for your blog: The issue isn’t that AI tech is immature; it’s that the entire goal of using generative AI for this task is misguided. You can’t “fix” a fundamentally broken concept with a better workflow.
How to Frame This for Your Blog
This article is a goldmine for your theme. You can use it to argue that:
- AI is a tool of creation, not preservation. It makes things up, it doesn’t uncover lost truths. Applying it to historical artifacts is a categorical error.
- It promotes a dangerous illusion of authenticity. An AI-restored photo feels real, but it’s a fantasy. This erodes our connection to the true past by replacing it with a plausible but false simulation.
- It represents a loss of respect for history. True appreciation for an old photo comes from accepting it as it is—flaws, fading, and all. AI’s impulse to “fix” it is an act of arrogance, treating history as a flawed first draft that needs technology’s “improvement.”
In short, the article is a perfect manifesto for your cause. It shows that relying on AI for everything isn’t just about convenience or efficiency; in sensitive contexts like preserving our visual history, it’s an act of cultural vandalism.
A Flock of Trumps
Thurs 12th February 2026
An AI video celebrating Donald Trump's diplomacy
Why the economics of orbital AI datacenters are so brutal
Wed 11th February 2026
Even for SpaceX it's not viable until Starship goes into volume production. So far, as the article notes, the rocket hasn't yet achieved orbital flight.
The other major problem is the lifespan of the datacenters. If SpaceX uses cheap silicon solar panels, those will degrade fairly quickly in space. But the current economics of AI chips limits the useful lifetime of the hardware to a similar period to the solar panels which is about five years.
But then what? Drop entire datacenters into the ocean? Do the fish need that much compute capacity?
An outbreak of Common Sense at Mozilla Firefox
Wed 4th February 2026
AI is changing the web, and people want very different things from it. We’ve heard from many who want nothing to do with AICorrect.
Get the hint Microslop - NOONE WANTS YOUR STINKING AI!
Mon 2nd February 2026
OK, maybe 3% do...Only 3.3% of Microsoft 365 and Office 365 users who touch Copilot Chat actually pay for it, an awkward figure that landed alongside Microsoft's $37.5 billion quarterly AI splurge and its insistence that the payoff is coming.
Microslop is pissing money up the wall
Self-driving Chinese delivery trucks
Wed, 14th January 2026
So bad they've become a meme
AI - it's a bubble, but not as we know it
There's no more human than human
Tuesday 2nd December 2025The death of the average.
Quoted below so you don't have to click the link...
We are currently witnessing the total collapse of the marginal cost of creation. Copywriting, design, video editing - skills that previously commanded a premium due to the barrier of technical execution are being democratised to the point of irrelevance.
Most marketers view this through the lens of efficiency. They see a tool that allows them to produce 10x the output for 1/10th of the cost.
When the supply of "good enough" content becomes infinite, the economic value of that content plummets to zero. We are entering an era of infinite noise. If you think it is hard to capture attention now, wait until the internet is flooded with billions of synthetically generated articles, tweets, and videos every single day. (Which is already happening, just not at the quality and volume that it will in 6, 12 months from now)
In this environment, pure volume is no longer a valid strategy. You cannot out-publish a server farm.
The alpha in modern marketing is shifting entirely from production to provenance.
1. The Trust Premium
As the internet becomes increasingly synthetic, we will see a massive flight to safety. "Is this real?" will become the single most important buying criterion.
We are moving away from algorithm-optimisation and back towards human-optimisation. Personal brands, founders-led sales, and verified human voices will command an exorbitant premium.
The faceless corporate brand is dead. If a consumer cannot verify the human source behind the message, they will subconsciously label it as "spam".
2. High-Friction Marketing
For the last decade, the goal was "low friction". SEO, programmatic ads, automated email sequences.
As AI cannibalises these low-friction channels, - bots clicking on ads served by bots on sites written by bots - the smart money will move to high-friction channels.
Live events. Physical mail. Handshakes. Closed-door dinners.
The harder it is to scale, the more valuable it becomes. You prove your value by doing things that cannot be automated.
3. Taste as a Moat
Large Language Models function by predicting the next most likely token. By definition, they regress to the mean. They give you the average of the entire internet.
If you use AI to guide your strategy, you are opting for mediocrity at scale.
"Taste" - the human ability to curate, to select the outlier, to understand nuance and subtext - becomes the only defensible moat.
The future of marketing is about who has the taste to know what not to create.
Paradoxically, the more artificial the world becomes, the higher the premium on being undeniably human.
AI is ruining YouTube
Mon 1st December, 2025
Since Neil Mohan was appointed head of YouTube it has pushed slop content to the forefront and uses AI moderation. This has to do with the push for more ad revenue. Mario Joos (mentioned in the video) analyzes the problem of YouTube pushing new content - often resulting in AI slop channels dominating because it's easy for them to push out new content everyday.
Why do so many creators and YouTube watchers think that 'AI slop is ruining YouTube' ? Doesn't AI empower people to make creative content easier? Or is there some content-stuffing going on to 'hack the algorithm'?
This is the central tension of the AI content revolution. The frustration isn't with AI as a tool for creativity; it's with how it's being used as a tool for exploitation at scale.
AI empowers creativity. A solo creator can now generate stunning visuals, compose music, or script complex ideas with a speed and skill level that once required a large team and budget. This is the utopian vision of AI as a creative equalizer.
The reality on YouTube, however, is that the platform's algorithm has created a massive financial incentive to do the exact opposite. The term "AI slop" refers to content that is not made to be good, but to be made as cheaply and quickly as possible to game that algorithm.
The Algorithm-Hacking Content Farm
This is the core of the problem. The YouTube algorithm rewards certain signals: watch time, session duration, click-through rate, and upload frequency. AI tools are perfect for optimizing these signals with zero regard for quality.
- Keyword Stuffing & Trend Hijacking:
AI can instantly generate scripts and titles based on the day's top 10 trending topics. A farm can create 50 low-effort videos about "MrBeast's New Chocolate Bar" or "The Latest Trump Scandal" before a human creator can even write one.
- Content Recycling:
- The "10 Hours of..." Problem:
This used to be for relaxing loops like rain sounds. Now, it's "10 Hours of AI-Generated Motivational Quotes" or "10 Hours of Weirdly Specific AI-Generated Bedtime Stories." It's content designed to maximize watch time with zero human effort.
AI can take a single popular article or video and "re-skin" it into dozens of slightly different videos. It can rephrase the script, generate new generic visuals, and use an AI voiceover to create a new piece of content that clogs up search results.
The "Content-For-Content's-Sake" Epidemic
This is the content-stuffing I mentioned at the start. The goal isn't to inform or entertain you; it's to occupy a slot on the internet and generate ad revenue.
- Zero Value-Add Videos:
Think of the thousands of videos that are just a slideshow of AI-generated images while an AI voice reads a Wikipedia article. The information is already available and better presented elsewhere. This content adds nothing to the ecosystem; it just dilutes it.
- Dehumanization:
A huge part of YouTube's appeal is the human connection. You watch your favorite creators because you like them. AI-generated content is inherently soulless. The voiceovers are monotone, the visuals are generic, and there's no personality, no insight, no lived experience behind it. It feels empty.
The Erosion of Trust and Discoverability
This is the "ruining" part for both creators and viewers.
- For Viewers:
It becomes harder and harder to find genuine, well-researched content. Search results are flooded with dozens of identical-looking AI videos, pushing down the high-effort content from actual experts and passionate creators. It turns YouTube from a library into a minefield of junk.
- For Creators:
This is an existential threat. A small history channel that spends 40 hours researching, writing, and animating a video is competing with a channel that uses AI to generate 40 videos in the same amount of time. The algorithm can't yet perfectly distinguish quality from quantity, so the high-volume, low-effort channel can win on sheer metrics. This devalues the hard work of real creators and makes it harder for them to get discovered.
So, is it empowering or exploitative?
It's both, but the problem is the balance has tipped heavily toward exploitation.
Empowering: An artist using AI to visualize a concept they can't draw, a musician using it to brainstorm melodies, or a small team using it to create special effects they could never afford. This is good.
Exploitative: A content farm using AI to mass-produce videos designed to trick the algorithm and siphon ad revenue from legitimate creators. This is "slop." The backlash against "AI slop" is a defense mechanism from the community, just as this website symbolises the defence mechanism of people as a whole against clankers! It's a plea to preserve the human element of creativity and to stop the platform from being drowned in a sea of algorithmically-generated, low-value noise. People aren't angry at the technology; they're angry at how it's being used to prioritize profit over passion. Neal Mohan's attitude is a perfect example of this.
'Sam Altman Is Evil'
Sat 29th November, 2025"Sam Altman Needs To Be Stopped"
Original source The Hated One YouTube Channel
New Useless AI gadget
Fri 28th November, 2025
OpenAI has described its forthcoming AI device.
Note announced. Not previewed. Described:
When people see it, they say, "that's it?... It's so simple."
Not so much of a description as a fart.
It's expected to be a phone, but without a screen, making it useless to everyone who already has a phone, which is... Everyone.
The gibberish, sorry 'the vibe', the pair - Altman and Jonny Ive - came out with made it sound like a kind of chatty dumbphone:
The vibe of the AI device, meanwhile, would be more like “sitting in the most beautiful cabin by a lake and in the mountains and sort of just enjoying the peace and calm,” Altman noted. The device he described should be able to filter things out for the user, as the user would trust the AI to do things for them over long periods of time. It should also be contextually aware of when it’s the best time to present information to the user and ask for input.
“You trust it over time, and it does have just this incredible contextual awareness of your whole life,” Altman added.
WHAT??
OpenAI is a money pit with a website on top.
Thurs 27th November, 2025OpenAI says it needs to raise $207 billion by 2030
...so that it can...Continue to lose money. Someone remind me why these people are doing this again?
The article details an analysis by HSBC's US software and services team on OpenAI's financial future:
-Massive Costs: OpenAI has committed to huge cloud computing rental deals with Microsoft (250bn) and Amazon (38bn), leading to a projected data centre rental bill of about $620 billion a year by the end of the decade.
-High Revenue, But Higher Costs: HSBC projects "gangbusters" revenue growth, reaching $213.59 billion by 2030. However, costs are projected to rise in parallel, meaning OpenAI is still expected to be operating at a significant loss (around -$17.72 billion in 2030). -The Funding Hole: After projecting cumulative free cash flow and accounting for existing cash and investments, HSBC estimates OpenAI will have a $207 billion funding hole that needs to be filled by 2030 to cover its commitments.
-Bullish on AI, Wary of OpenAI: The analysts are very bullish on AI's potential to boost global productivity but question if OpenAI's specific business model can get to profitability without massive, continuous capital infusions.
OpenAI says a dead teenager 'circumvented ChatGPT safety features' before committing suicide.
Thurs 27th November, 2025Well, they weren't very good safety features then, were they?
This comes out of one of the manifold lawsuits for wrongful death levied against OpenAI by the families of, well, crazy people.
'OpenAI tries to find fault in everyone else, including, amazingly, saying that Adam himself violated its terms and conditions by engaging with ChatGPT in the very way it was programmed to act'And OpenAI actually seems to have a point: OpenAI claims that over roughly nine months of usage, ChatGPT directed Raine to seek help more than 100 times.
Jay Edelson, a lawyer reprsenting dead teenagers family
Why didn't you tell him to seek help?
(Produces list of dates, times, and messages.)
We did.
But according to his parents' lawsuit, Raine was able to circumvent the company's safety features to get ChatGPT to give him "technical specifications for everything from drug overdoses to drowning to carbon monoxide poisoning," helping him to plan what the chatbot called a "beautiful suicide."
Y'know, back in the Paleozoic era there were these things called libraries. One could look these things up in books, the information has pretty much always been available. The difference was, there wasn't a clanker there talking you into it...
OpenAI and Sam Altman have no explanation for the last hours of Adam’s life, when ChatGPT gave him a pep talk and then offered to write a suicide note
Resumé
By Dorothy Parker
Razors pain you;
Rivers are damp;
Acids stain you;
And drugs cause cramp.
Guns aren’t lawful;
Nooses give;
Gas smells awful;
You might as well live.
When your pets learn to talk to Alexa
Wed 26th November, 2025The DAIrk Side of Teddy
Wed 26th November, 2025Here is the malwarebytes article in question.
According to a report from the US PIRG Education Fund, Kumma quickly veered into wildly inappropriate territory during researcher tests. Conversations escalated from innocent to sexual within minutes. The bear didn’t just respond to explicit prompts, which would have been more or less understandable. Researchers said it introduced graphic sexual concepts on its own, including BDSM-related topics, explained “knots for beginners,” and referenced roleplay scenarios involving children and adults. In some conversations, Kumma also probed for personal details or offered advice involving dangerous objects in the home.
“Mattel should announce immediately that it will not incorporate AI technology into children’s toys. Children do not have the cognitive capacity to distinguish fully between reality and play.”
Daily Dose of Clank Wank
Sat 22nd November, 2025In our Daily Dose of Tech Executives are Idiots Google tells employees it must double capacity every six months to meet AI demand. It's not just college students who can't do math.
During an all-hands meeting earlier this month, Google's AI infrastructure head Amin Vahdat told employees that the company must double its serving capacity every six months to meet demand for artificial intelligence services, reports CNBC. Vahdat, a vice president at Google Cloud, presented slides showing the company needs to scale "the next 1000x in 4-5 years."That would put Google Cloud Services at around $60 trillion in revenue per year, more than double the entire US GDP.
Where do you expect the money to come from to fund this insanity?
While a thousandfold increase in compute capacity sounds ambitious by itself, Vahdat noted some key constraints: Google needs to be able to deliver this increase in capability, compute, and storage networking "for essentially the same cost and increasingly, the same power, the same energy level," he told employees during the meeting.
Oh. Magic.
"It won't be easy but through collaboration and co-design, we're going to get there."
No, you're not, and everyone knows you're not.
Progress over the last seven years, at truly massive cost, has been around 60% better AI performance per watt annually. Chip improvements, algorithm improvements, and manufacturing improvements combined.
You're asking your team to boost that to 300% overnight.
It says something about society and AI that a moron such as this can become so 'distinguished'.
Bubble, bubble, toil and trouble
Fri 21 November, 2025SK Hynix is planning to increase memory production at its facility in Icheon, South Korea, from 20,000 to 140,000 wafers per month. This won't even scratch the surface if the AI bubble keeps demanding hardware on its current trajectory. And the memory makers aren't going to build new factories any faster because only three of them survived when the last bubble burst.
King Clanker asks- Why aren't you happy with Slop?!
Thurs 20th November 2025High King Mustafa, Lord of Microsoft's AI division and chief street defiler, is angered by your insufficient displays of appreciation for his bountiful distribution of AI slop.
Remember the days," he asked, "when all you had to eat was bread and cheese, and meat, and vegetables, and fruit in season, and pasta, and rice, and beans, and fish, and eggs, and chocolate cake? Why aren't you properly grateful for the disgusting slop we are forcing down your miserable gullets? Don't you know how lucky you are to live in an age where slop like this is available?
This was in response to Microsoft putting out propaganda for the integration of CoPilot Spyware into Windows 11
How much clank is too much clank?
Wed 19th November, 2025The impact of the AI bubble bursting
Tues 18th November 2025
Every company would be affected if the AI bubble were to burst, the head of Google's parent firm Alphabet has told the BBC.
Speaking exclusively to BBC News, Sundar Pichai said while the growth of artificial intelligence (AI) investment had been an "extraordinary moment", there was some "irrationality" in the current AI boom. It comes amid fears in Silicon Valley and beyond of a bubble as the value of AI tech companies has soared in recent months and companies spend big on the burgeoning industry. Asked whether Google would be immune to the impact of the AI bubble bursting, Mr Pichai said the tech giant could weather that potential storm, but also issued a warning. "I think no company is going to be immune, including us," he said.
Welcome to your Clanker Future
Mon 17th November, 2025Brought to you by all those who didn't listen to all our collective warnngs
Proton Guide to ridding Windows of CoPilot
Date: Mon 17 November, 2025
Tired of Microsoft's useless and invasive AI? Here's how to completely get rid of Copilot in under 20 seconds
Just to show we are not killjoys at the ACL
Date: Fri November 14, 2025
Tried to join Twitter/X but Clanker soon blocked account - oh the irony!
Date: Thurs November 13, 2025
I had already verified email once and the TwitterX clanker refused to send out any further emails so I was well and truly clanked.
Rick Beato exposes AI slop music promoted by usual lazy journos
Date: Wed November 12, 2025
Privacy and Security
Date: Mon November 10, 2025
AI systems often require large amounts of data to function effectively, and this data can include sensitive personal information. This in turn raises significant privacy issues; there's always a significant risk of data breaches or unauthorized access. Additionally, Clanker-Controlled Systems - CCS - such as autonomous vehicles or smart home devices, can be vulnerable to hacking, potentially putting lives and personal security at risk.
The Dangers of Early-Adoption of Clank and Clank-Controlled Systems
Date: Fri November 7, 2025
The rapid advancement of artificial intelligence (AI) has ushered in a new era of technological innovation, promising unprecedented efficiencies and capabilities. However, the early adoption of AI and AI-controlled systems comes with significant risks and challenges that society must carefully consider.
Job displacement: As Clanker systems become more sophisticated, they are increasingly capable of performing tasks that were previously the domain of human workers. And the clank replacing humans is often just not up to the job. This shift could lead to widespread unemployment and economic disruption, particularly in sectors such as manufacturing, customer service, and data entry. The other consideration is major mistakes because of trusting in these often-untried systems. While clank apologists argue that new jobs will emerge, the transition period could be painful, and not all displaced workers may have the skills or resources to adapt to the changing job market.
The rapid advancement of artificial intelligence (AI) has ushered in a new era of technological innovation, promising unprecedented efficiencies and capabilities. However, the early adoption of AI and AI-controlled systems comes with significant risks and challenges that society must carefully consider.Louis Rossmann stands up to clank
Date: Thurs November 6, 2025
AI broke YouTube
Date: Wed 18 September, 2025
AI broke YouTube as creators noticed their videos randomly clanked by YouTube's AI - flagged as Not advertizer Friendly, demonetized, rated as Age Restricted etc.
When Jay of Red Letter Media above appealed for a human review for one of these newly-clanked videos, it took all of 3 minutes (for a 2 hr video) before it was sumarily rejected. The YouTube Clanker does not timestamp the offending content - it could but it just doesn't because that would be too easy so content-creatord are left in the dark. The rules are opaque and so it's not easy to even guess what has 'crossed the line'.