Thanks to visit codestin.com
Credit goes to cubicgarden.com

Tag: ai

Public Service Internet monthly newsletter (Jan 2026)

ChatGPT psychosis? This scientist predicted AI-induced delusions — two years later it appears he was right This scientist predicted AI-induced delusions…

We live in incredible times with such possibilities that is clear. Although its easily dismissed reading Arduino’s new terms of service, reading the awful story of the man behind evil twin wifis on airplanes and please stop using AI browsers.

To quote Buckminster Fuller “You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”

You are seeing aspects of this with Apple working with the EU’s DMA, E-petition digital ID debate now on youtube and UK flat owners could have a legal right to fibre.

France turns hostile to privacy and encryption?

Ian thinks: Worrying that the stance on open Android apps outside the play-store is undecided, but even worst to see the crack down on encryption comes to the likes of privacy centred Graphene OS.

Remember the pipe dream of the metaverse?

Ian thinks: Travel back a few years when Facebook become Meta, Horizon worlds was on the tip of lots of tongues. But for the wrong reasons as this video makes super clear. Even I was shocked to see an average of 900 people using it along side #4.4 billion in loss income while Zuck’s glaze is elsewhere.

Remember Facebook dating?

Ian thinks: Talking about Meta, surprisingly Facebook dating its doing very well, although not promoted or talked about at all. There’s a plan to do more in 2026, I expect the changes at Hinge will have an influence to reflect Zuckerberg’s plans for more AI bots?

Dating coaching with Chat GPT?

Ian thinks: We all know people who are using AI for different aspects of life but as a dating coach is on the up-tick  The notable part is this has been published in the female fashion magazine Elle, and indicates a worrying trend of AI mingling with human intimacy.

The enshittification of Airbnb and so much more…

Ian thinks: As a Airbnb host myself, I can agree about the enshittification of the platform Airbnb. I personally have had to fight against the decay of the platform from AI pricing, scramble for superhost status and so much more.

The everyday problems of privacy

Ian thinks: This video explains just a few of the problems with being private on the internet today. I recently did similar and found it painful, emailing back and forth to confirm everything, in the end I just gave up.

Talking LLM bias in detail

Ian thinks: AI bias was heavily talked about for years, then things went quiet. I found this detailed conversation with Harvard psychologist Mahzarin Banaji, perfectly timed in the face of the rise of LLMs. I found the API vs chat interface differences really powerful.

AI psychosis? What could be done?

Ian thinks: You likely have heard of it but its when AI echos and encourages paranoia and delusions. There are many real examples, but this well watched video although curated for performance shows how things can accelerate downwards ever so quickly. The big question is what can be done?

The authoritarian pipeline made visible

Ian thinks: Francesca Bria and others clearly highlight the huge problems with the current tech setup in the states. In this interactive mapping, its insane how interwoven everything is, but hardly surprising. What this means for Europe is a big question and the key point.

The next 10 years of public service in the UK?

Ian thinks: Every 11 years the UK government puts out a green paper for the future of the BBC public charter. As a public consultation, there is a call for feedback and thoughts on the future of public service and the BBC. Highly recommend spending the time on this, and its open to all

Like this newsletter? Find the monthly archive here

Public Service Internet monthly newsletter (Dec 2025)

Traffic jam of cabs in New York shot from the back with red lights from the cabs
Cars everywhere: (cc-by-nd) https://www.flickr.com/photos/strolicfurlan/24747443097/

We live in incredible times with such possibilities that is clear. Although its easily dismissed seeing the questionable AI courtcase in the UK, Zuck running an illegal school and big tech refusing to talk about election influencing in the Netherlands.

To quote Buckminster Fuller “You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”

You are seeing aspects of this with Android supporting airdrop, The Eurosky sovereign bluesky project now has open code and Dr Ruha Benjamin’s unlearn speech from Mozfest 2025.

The future of addiction

Ian thinks: I found this conversation spot on, as it touches on many things I find really interesting right now and links them up. From the problems of mass social media to weaponized addiction being generated on mass. Addiction is always a contentious point but regardless, it all points to a future where our revealed preferences are indulged without friction for profit and control..

Why we need a war on cars

Ian thinks: I don’t drive a car, and this episode is full of interesting points including the one about superman, electric cars effect on space and the city and the social impact on loneliness.

The characteristics of a AI bubble

Ian thinks: If you are in the tech field or even looking in, its clear everybody is talking about the AI bubble but this podcast from Wired again, really breaks down the characteristic of a bubble.

Keep Android open

Ian thinks: Great site, seems Google might have listened to the pressure in recent days. What they mean by an alternative is a big question however.

Reasons for divorce: AI relationship

Ian thinks: We all new this was coming. You message first thing in the morning, last thing at night, you message on the toilet! That relationship seems so innocent but engaging. Now there is legal challenges a foot, which could make that relationship grounds for a divorce.

Digital legacy and death tech overview

Ian thinks: This video from Wired magazine, starts off with the usual cryogenic nonsense  but then grief bots and AI avatars suddenly become a lot more real. I especially like the shout out to get a will and manage your digital legacy. This is something I am working on with a project called Machine readable wishes.

NSA’s mobile phone guide

Ian thinks: The NSA’s mobile guide is interesting reading, there are nuggets of information such as restart your phone weekly and have a case which block room audio. Paranoid or just good advice?

Bots are the future of social media?

Ian thinks: I recently gave a talk about AI to a health insurances. I talked a lot about loneliness and bots. Although I didn’t include it, this detailed video forecasts the logical end of social being the dead internet theory but far far worst.

What does 47,000 conversations with ChatGPT tell us?

Ian thinks: Remember when OpenAPI was sharing chatGPT conversations? Well analysis of the conversations by the Gerrit De Vynck tells us exactly how puppy like and integrated it is with peoples lives. Counter to what the AI companies have been saying, this is worry-some to hear and read.

Like this newsletter? find the monthly archive here

Best tips for dealing with your digital legacy

Wired magazine has a good summary of the death tech or I prefer digital legacy space. I can believe the sector is worth  125,000,000,000 dollars but people are throwing money around in the wrong places I feel.

The best part of the video comes when talking about what people can actually do to take some control over their digital legacy. First step get a will, talk with your family and finally take some control of your data now.

This is why I have been working on machine readable wishes, which work alongside a will. It could be super useful to help people take control of their legacy in the future. The beauty of it is, that it is legal now (in common law countries), doesn’t require any major changes and is open source.

To learn more check out the link site, blog, slides explaining the project and the github code which works with N8N.

Public Service Internet monthly newsletter (Nov 2025)

This is wearable maximalism but subtract two smartwatches and smart glasses, and this could be you in three years.
From the Verge

We live in incredible times with such possibilities that is clear. Although its easily dismissed seeing the deliberate thoughtfulness of Tilly Norwood,  Deloitte’s AI report to the Australia government and Windows 10 end of life is a e-waste disaster?

To quote Buckminster Fuller “You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”

You are seeing aspects of this with Sir Tim’s thought on the future of AI,  renewable overtaking coal as the worlds source of electric and digital sovereignty grows bigger in Europe.

The future of cities: Humans are the problem?

Ian thinks: Recently I attended the Society 5.0 conference and attended a workshop on how the internet is changing architecture and physical public spaces. At the same time by chance McNeil and Marx talk through the same tensions.

AI slop, rage bait, whats next?

Ian thinks: This discussion between Taylor and Drew is very revealing, how prolific Ai slop really is. Its a detailed discussion and with lots of in-depth commentary on the process of how its made and why.

Lets talk about the AI bubble

Ian thinks: Everyone is talking about the AI bubble but this video breaks down what a bubble looks like and with a critical eye how money is being passed around. How this works and are we really looking at the collapse of the last decades?

Deep insight into Europe’s digital mindset

Ian thinks: During the PublicSpaces conference earlier this year, there was a great keynote talk in Dutch but a big part of the talk was around European culture and Startup culture clash. The German Digital-Mittelstand is a strong example of the long-term view of the European landscape.

Bluesky federation isn’t ready?

Ian thinks: Bluesky has been talking about federation for a long while but its clear over this high profile this conflict with Blacksky.
Futher conflict is around the corner with the Trump administration’s arrival on Bluesky.

Smart glasses, or work for Ai?

Ian thinks: The emphasis on smart glasses recently has raised a ton of red flags. It feels like there is so little learned from the past including Google glass. Gilliard and Paris break down what I have been thinking how little has been learned. On a more casual note.

Enshittification and so much more from Cory

Ian thinks: Cory has been on the circuit talking about his new book Enshittification. I have heard a few interviews but one of the best is certainly the one with Rushkoff where they talk about so much more. Second best is this one on Vox’s grey area.

The Intentional Collapse

Ian thinks: Talking of Douglas Rushkoff, he’s monologue on a potential intentional international collapse due the Ai bubble, greed, quality and so much is a sobering reminder of what’s at stake. But also the small steps we can do to make a difference.

Straight out of America a man called Mark Bray

Ian thinks: You may have come across this story about Mark Bray and his radical works casing him and his family to be put on a watch list of radicals. It’s a story in a worrisome trend of researchers joining such lists along side journalists and activists.

Amazon’s outage says more than you can imagine

Amazon goes down and so many of the services we rely on daily and there are many thoughts about that? However there is so much more to this story than first imagined. This video explains how deep this all potentially runs.

Like this newsletter? Find the monthly archive here

Public Service Internet monthly newsletter (Oct 2025)

Mesh network running off solar on top of a house

We live in incredible times with such possibilities that is clear. Although its easily dismissed seeing AI attempting to rewrite history, Agentic AI browsers fooled and Rabbit R1 is back?

To quote Buckminster Fuller “You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”

You are seeing aspects of this with good counter-points parts [1][2][3] to tell the AI die-hards, a uptake of LoRaWAN and low energy decentralised networks thanks in part to meshtastic and alternative to typical internet servcies but European based look no further.

Small acts of community interest

Ian thinks: Rushkoff’s story about borrowing a drill and community has become similar to the billionares and the event story. Rushkoff calls it a revolutionary act, I call it an act of community interest, something we rarely see in

The case for no billionaires

Ian thinks: This interview with professor Ingrid Robeyns, is a good listen especially after reading Limitarianism and thinking about the problems of inequality. I wonder if anyone in government is taking notes as there are so many voices and credible research pointing to the same conclusion.

Right to repair hard truths

Ian thinks: Like it or not, Rossmann speech about the right to repair, how we can’t own anything as the makers find more clever ways to enshittify is on point. I am sure Samsung’s fridge with adverts you can’t turn off, would be in a updated version if redone now..

Empires and their religions they build

Ian think: Karen Hao, has been the rounds recently with the Empires of AI book. This is one of the best conversations/interviews I seen with her. The mid point and end points are well argued by Hao, touching deeply on the need to make their own religious pyramids of technology.

Do your friends still post on social media?

Ian thinks: Do you remember when people use to “hard” post? What happened? According to Chayka many things. From algorithms, ephemeral, context collapse, advertising, etc. Let’s not forget the effect of bots too.

Learning from the Ozone crisis

Ian thinks: in 1985 most won’t remember the Ozone hole threat as its so baked into the world we live in now. Tristan talks with Susan about the solution and how it can be applied now to the AI challenge.

Ultra-processed authoritarian algorithms

Ian thinks: Nobel peace-prize winning journalist’s conversation with the Daily shows’s Jon Steward about the state of journalism and authoritarian, is a rare and frank exchange with Maria Ressa making it super clear how the playbook happened almost a decade ago and how they tried to alert the world back then.

What is the difference between Android and iOS?

Ian thinks: If this is true and the video summary isn’t off the mark. There is a big question about where do people go? We always knew having a duopoly for mobile devices was going to blow up in our faces.

AI’s branding team are on fire

Ian thinks: Meredith Whittaker tells it as it is, is great in this interview about how she got started in the tech industry, signal gate, AI, masculinity and so much more. Her insight on how “useful” AI agents are infiltration services such as Whatsapp and Signal is spot on.

Like this newsletter? Find the archive here

Public Service Internet monthly newsletter (Sept 2025)

Web browser with a Age gated website shown

We live in incredible times with such possibilities that is clear. Although its easily dismissed seeing how we are being watched, wishing people will stop sharing pictures with ChatGPT and people in the UK told to delete emails In order to save water?

To quote Buckminster Fuller “You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”

You are seeing aspects of this with lessons for the AI future in Supremacy, the real need for deniable encryption and how strong 3D printing has got.

You know its bad, when RSS is all you look forward to?

Ian thinks: This discussion between Molly White and Ed Zitron, RSS is mentioned as a really good example of a technology which is reader/person friendly and is a great example of how its simplicity and ecosystem is a example of what we could all learn from.

Has the Roman Church’s view on AI changed?

Ian thinks: In this podcast Paolo Benanti, digs deep into the human and shared society values of many of us, in face of the silicon valleys dreams of AI solving all. I especially like how putting aside differences to work on the bigger problem, is actually working.

The office wars are back?

Ian thinks: As someone who wrote XML stylesheets to convert Microsoft Office XML in 2003 to xHTML and PDF. I can completely understand Libreoffices deep concerns and how the old Microsoft mentality of embrace and extend, lives long

What can we say to grads entering the difficult job market

Ian thinks: There has been much said about AI taking jobs, well its happening but in ways not expected. Of course its not just AI, but there is a whole wave of different concerns causing the real difficult discussion with fresh new grads

Windows 12 sounds like a true nightmare?

Ian thinks: If this is the future of Windows is voice first and AI everywhere you turn. Its clear Microsoft vision of Windows is a privacy nightmare. More so than any other operating system currently used. Will it encourage people to jump? Unlikely, sadly.

A glimpse of the web we don’t want?

Ian thinks: Talking of nightmares, many of us have used the wayback machine at some point. However this is the way-forward machine, giving a spicy glimpse of where we are heading with the web if things move in the same direction. A warning from the future if you haven’t been paying attention…

Leaking data and how AI could social engineers us?

Ian thinks: This intriguing long conversation starts with social engineering and how social engineering with the data we share and trust in others. Then turns towards AI and the threat of engineering from AI systems.

Do countries have true commitment to sovereignty?

Ian thinks: Cecilia Rikap’s open remarks about government and the public interest in regards to the UK speaks volumes. Then turning to Europe using parallels of South America and data colonising is spot on. Are countries willing take sovereignty for real or are the cracks growing.

Why is piracy on the rise again?

Ian thinks: I tried to find a good summary of the wider scope of whats happening (as many are focused on costs alone) but the best I can find is this video from moon. Private equity flatters everything and is something driving more enshittification.
Of course no one is condoning piracy but the times are changing?

Like this newsletter? Find the monthly archive here

A lesson in separation of concerns: Google’s gemini hijacked

The news Google’s Gemini AI was hijacked via a hidden prompt in a calendar invite.

Hackers Hijacked Google’s Gemini AI With a Poisoned Calendar Invite to Take Over a Smart Home. For likely the first time ever, security researchers have shown how AI can be hacked to create real-world havoc, allowing them to turn off lights, open smart shutters, and more.

Rings bells to me about the lack of human data interaction especially the separations of concerns.

I already talked about this with Google’s AI wanting access to everything Google related. This hijack wouldn’t be so effective if there were better permissions and separations of concerns were built in rather than given lip service.

Google sent an email today with the title… Introducing temporary chats and new data controls. Still not addressing the underlying problems.

When will they learn???

It really makes me think about a few things.

Will Android fork off sooner or later? Will moat people care?

Pipelines and AI?

A long time ago I described pipelines and where they could go.

I was made aware of Google’s OPAL project and reminded of how strong pipelines can be for managing complexity.

The pipes are described by prompts but interestingly you can reuse and share pipes and the whole flow. Which leads me to wonder if there might be a little market place for the pipes?

Might need to keep an eye on Opal, but wondering if there are better solutions. For example Xproc which can interface with APIs of large language model without getting into the weeds?

Public Service Internet monthly newsletter (Aug 2025)

2 people sat in chairs looking at the camera across a table with a cup on it
Black Mirror 7×1 Common People (2025)

We live in incredible times with such possibilities that is clear. Although its easily dismissed seeing browsers extensions turned into scraper bots, bodies used to fingerprint you with wifi and Mozilla under more pressure.

To quote Buckminster Fuller “You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”

You are seeing aspects of this with Duck duck go hiding most AI generated images in searches, looking at Proton’s Lumo privacy first AI and some children naturally reducing their phone to manage their mental health

Imagine if all phones were built this way?

Ian thinks: The fairphone has been around for years but version 6 of the smartphone, seems to have got everything right. This tear down by iFixit, shows exactly how easy it is to repair, replace and even upgrade parts in the future. One day all phones will be built this way?

Welcome to the age gated internet?

Ian thinks: Age verification has hit the UK and there is a lot to say. The VPN usage increase the UK in the UK speaks volumes. The UK isn’t the only one but the impact is being felt by everyone. If the recent Tea app and so many other data breaches has taught us, sharing personal data is not going to go well. I’m hardly hearing ZKP (Zero-knowledge proof) mentioned, which is good practice for this.

Degrowth is growing?

Ian thinks: There are many mentions of de-growth dotted across these notes over the last few years. This BBC News video gives a reasonable view of the movement, some of the challenges and best of all calling out some of the presumptions which plague the movement.

Cloudflare may have cut off 99% of AI crawlers?

Ian thinks: Its been well talked about and if you own a site, you may have noticed the huge amount of traffic caused by bots. Cloudflare have had enough and wrote a open solution called Anubis which is too expensive for AI bots to get around.

Not the last word on AI Slop

Ian thinks: You know its really bad when John Oliver spends most of his weekly show talking about the problem and how its affecting everyone. Leaning on the work of 404 media, Oliver uses comedy to great effect pointing out the insanity of were we are now.

Cory and Maria in conversation

Ian thinks: As part of the Open Rights Group’s celebration of 20 years since its inception on that day in London during the Open Tech conference. The conversation doesn’t sit in the space of nostalgia too long, but rather looks at what was learned and how it can be applied in current time.

The big questions around human brain interfaces

Ian thinks: There has been an uptick in Human brain interfaces news and discussion. However few are asking the really big questions we all have about the technology. Thankfully Coldfusion is asks theose questions about privacy, enshittification, agency etc. Dare I mention Black Mirror’s common people?

Decomputing AI for a better future?

Ian thinks: The critique of AI is well thought-out by McQuillan, and the relation to the context collapse, neo-liberalism and empire building is spot on. The ability of the AI empires to fill in the narrative when there isn’t one is also a key point. Well worth reading Empire of AI too for related impact, which I just finished.

Did you recently die or was that obituary fake?

Ian thinks: Is there no line AI slop won’t cross for the clicks? Fake obituaries are appearing across the internet and its another example of the slop and it is pretty bad news for its target/victims, the friends and family members. The video is also worth watching too.

Like this newsletter? Find archive here

Public Service Internet monthly newsletter (July 2025)

A couple look at a TV while the camera looks over their shoulders

We live in incredible times with such possibilities that is clear. Although its easily dismissed seeing how how much time we are spending on our smartphones. whats happening with GDPR and the low-background steel problem of the AI contaminated internet?

To quote Buckminster Fuller “You never change things by fighting the existing reality. To change something, build a new model that makes the existing model obsolete.”

You are seeing aspects of this with scientists proving ethical sourced AI is possible, plastic which dissolves in water and opensource coming to rescue owners of the humane pin.

Where is the future of social web going?

Ian thinks: Its finally available, the report myself and BBC R&D colleagues wrote on the emerging crisis for the social platforms. Exploring the toxic and polarised online environments but also the enormous potential to transform the world in positive ways, if built with the right values.

Theranos is back?

Ian thinks: Remember the whole testing a drop of blood to diagnose critical diseases? Well believe it or not Holmes is back in the form of Haemanthus who’s CEO happens to be Holmes partner.

Oversharenting has its own murky marketplace

Ian thinks: I did see the Ella campaign a while ago, if you haven’t see it go watch it. The video from NBTV, goes much further looking into sharing data without understanding or concern of how others will abuse it.

Ian thinks: Automatic Content Recognition, is actually old news (as the UCL study shows)  but recently become a long clearer recently as more researchers and journalists take a look at the privacy policies of smart TVs.

Building the EuroStack

Ian thinks: There has been a lot of discussion recently about Europe as the 3rd way (between commercial and government). Digital sovereignty is pinned to the stack along with the move to strategic autonomy and being Europe its a  of the EU but isn’t exclusive to Europe. Providing

This is the year of Open source in Europe, doubtful?

Ian thinks: Great to see more European countries using open source but there is a big question… What went wrong last time? We assume technology would save us and didn’t. We need to trace back the power and money.

Love with AI, all dopamine and no rejection?

Ian thinks: Cathy Hackl, took the plunge into the AI as a partner, and the results are what you imagine but the gentle responses do connect on some level. Hackl does point to this being a good simulator, which I’m less sure of because we all know the business model behind them?

Could chatbots have a public service remit?

Ian thinks: Following the endless amount of chatbot news and last months in joke about the sentient machine therapist. Tiku and Marx, discuss the real harms of chatbots generally. Which leads me to think could a LLM based chatbot be trained and given the metric of public service. Dare I say the loneliness epidemic is one good reason.

Bounce your followers around?

Ian thinks: As seen at the Fediforum last month, Bounce brings the ability to move around federated social networks but keep the followers. This is something which makes the fediverse unique.

Like this newsletter? Find the archive here