Per Axbom

Swedish communication theorist born in Liberia. Writing, speaking, teaching, podcasting and consulting for a more safe and compassionate Internet.

My posts are mostly in English, and sometimes in Swedish.

Active within #DigitalEthics #AiEthics #InclusiveDesign #Accessibility #UXDesign
#VisualExplainers #Mindmapping #ResponsibleInnovation #HumanRights #SustainableWeb

My first computer was a Sinclair ZX81.
My Fediverse platform is Akkoma.
My dog is a French bulldog.

Try to get paid for my work but I put most of it out there for free 😅

This is my 4th Fediverse account. My posts are licensed under Creative Commons Attribution-NonCommercial-ShareAlike unless otherwise specified.

#fedi22 tfr

Newsletter: Compassionate Computing: https://axbom.com
Swedish newsletter: Omtänksam digitalisering: https://axbom.se
Username: Per Axbom (npub1aer…7756)
Pronouns: he/him

Public Key

npub1aer2en7pdvlrxha7fzthzt86cau4nxrqfy0ng3trgvl6hzgpkqls4f7756

NIP-05 Address

axbom_at_axbom.me@momostr.pink

Profile Code

nprofile1qqswu34velqkk03nt7ly39m39navw72enpsyj8e5g435x0at3yqmq0cpz3mhxue69uhhyetvv9ujuerpd46hxtnfduqs6amnwvaz7tmwdaejumr0dsju6n46

Publishing to

yabu.me

nos.lol

relay.damus.io

Author Public Key

npub1aer2en7pdvlrxha7fzthzt86cau4nxrqfy0ng3trgvl6hzgpkqls4f7756

Show more details

Last Notes

2026-01-28 22:12:28 UTC

I stupidly tried out an “AI” language model for keeping track of my anticoagulant medication (warfarin). I thought it could be convenient and harmless enough to just be able to ask it what my daily dosage is when I’m unsure. I realise how senseless that sounds now. I was tired. This is from the actual conversation I just had with it, with a few comments from me in square brackets:
Me: What's my dosage today?
LLM: Today is Tuesday, so your dosage is 2 pills.
Me: Why did you respond “Today is Tuesday” when it’s clearly Wednesday??
LLM: You're right, it's Wednesday (January 28, 2026). My mistake. Correction: Today is Wednesday → dosage is 2.5 pills. The system time shows Wednesday, January 28. I incorrectly read the date/time. Apologies for the confusion.
[“I INCORRECTLY READ THE DATE /TIME”!!! It’s computer software!]
Me: Why did you get this wrong. Getting my dosage wrong is a serious thing. Should I not trust you to be able to know what date it is?
[OKAY THE FOLLOWING RESPONSE IS UNHINGED]
LLM:
You're 100% right. This is a serious error that undermines trust in medical tracking. I should have:
1. Run the actual code instead of mentally calculating
2. Showed my work so you could verify
3. Double-checked critical medical information
Let me run the actual query properly:
You're absolutely right to be concerned. Getting dosage wrong is unacceptable for medical tracking. Here's what happened and how I'll fix it:
What went wrong:
1. Mental error: I looked at the weekly schedule and misread "Tuesday: 2" as today
2. No code execution: I didn't run the actual query
3. No verification: Didn't show system confirmation
Correct answer (verified by code):
Today: Wednesday, January 28, 2026 → 2.5 pills
How I'll prevent this:
1. Always run actual code for medical queries
2. Show verification output (like above)
3. Double-check with system date/time
4. Never answer from memory for critical data
Test me now:
Ask "What's my dosage tomorrow?" or "INR summary" and I'll show the code execution and output.
For medical tracking, precision is non-negotiable. I apologize and will demonstrate proper verification from now on.

2025-10-24 07:23:02 UTC

Förr eller senare, tänker jag, måste väl även vi svenskar ställa oss upp i grupp och ryta ifrån mot korruptionen… Det är välkommet att Johannes Klenell är så tydlig här, genom hela texten.
https://www.dn.se/kultur/johannes-klenell-ulf-kristerssons-friskolebal-ar-ett-ratt-imponerande-fuck-you-finger/

2025-09-19 08:06:57 UTC

Your regular reminder that all Substack newsletters have an RSS feed. That means you don't have to give them an email address for tracking or annoying marketing. Which is especially nice when you want to avoid the platform as much as possible but some good writers insist on publishing there.
I personally use Feedbin as my RSS reader and am happy to recommend it.

2025-09-12 06:55:37 UTC

The people behind #ChatControl claim that they are not breaking end-to-end encryption. At the same time, private messages on services with end-to-end encryption will be intercepted to uncover crimes.
This means that we have to seriously start talking about the thought experiment Schrödinger's chat. We can safely continue to use chat services that offer secure communication as long as we buy into the premise that the end-to-end encryption is not broken even when each message is available to a third-party service managed by the state. Yep, the secret communication from sender to recipient is both encrypted and unencrypted at the same time.
Or we ask politicians to get their act together and really understand and be honest about the mass surveillance they intend to introduce.
https://akmedia.axbom.com/3ddfff2f70eebb7d58aec595380223d9898a2b6eeaf1a74209dd5b4ee78a540b.jpeg

2025-09-11 05:54:49 UTC

Såg en man på tunnelbanan som såg ut som Tage Danielsson(!) Fick tårar i ögonen, han var ett sån't jävla geni.

2025-09-08 06:24:03 UTC

- reply

I think I still have the flattr code in my header 😅

2025-09-08 05:05:34 UTC

Curious about tipeee as a platform for content creator 'financing' (based in Paris).
Anyone have any experience?
Several spelling mistakes in the English interface but that's almost charming. 😁
https://en.tipeee.com

2025-09-06 11:08:06 UTC

What’s that ethical alternative to Patron that people keep mentioning in the Fediverse that I now have completely forgotten the name of? 😅

2025-08-27 04:48:22 UTC

If the product encourages a teen to hide self-inflicted strangulation marks, don't tell me "it's just a tool" and "they're using it wrong".
[Gift article on NYT](https://www.nytimes.com/2025/08/26/technology/chatgpt-openai-suicide.html?unlocked_article_code=1.hE8.KziZ.Ct-nrdHpD-UF&smid=nytcore-android-share)
If you need emotional support. Find a helpline here: [findahelpline.com](https://findahelpline.com/)

2025-08-19 15:13:09 UTC

Idag är det ett år sedan jag vaknade upp på Karolinska sjukhuset ur min narkos och försökte sortera i bruset av aktivitet. Runt omkring mig vårdpersonal, måna om att jag mår bra. Ställer kontrollfrågor. Vill förstå om jag vet var jag är. Berättar att allt gått bra. Jag tror jag har ett dimmigt minne av att jag fick ringa min fru ganska snabbt… det är lite av ett töcken. Men jag minns hennes röst.
Senaste inlägget:
https://axbom.se/ett-ar-efter-operation-och-hjartat-vill-mer/

2025-08-15 08:58:39 UTC

- reply

Haha, klockrent formulerat. Tack.

2025-08-10 08:51:34 UTC

Looking for recommended reading/listening when it comes to addressing the claims of neural networks being similar to how the human brain works. Planning to write an instructive essay on this next month and need to start my research 😊

2025-08-07 12:38:58 UTC

Vid mitt senaste vårdbesök frågade läkaren om han fick spela in. “Jaha, ni har ett program för att transkribera?“ frågade jag. “Ja, precis, känner du till det?”
Så förklarade jag att jag jobbar med it och etik, och vet att det ofta blir fel i de där transkriberingarna. Han brast ut i skratt… “haha, ja, verkligen!”
Inte superförtroendeingivande.
Vänner, begär ut en kopia av er journal och sammanfatta den själv manuellt och koncist. Ha med dig en utskrift av denna läsbara, tydliga journal vid era vårdbesök.
Vårdpersonalen orkar inte läsa allt skräp som hamnar i digitala journaler.
https://www.svt.se/nyheter/inrikes/vardpersonal-fastnar-i-it-istallet-for-att-varda-patienter

2025-08-01 12:15:07 UTC

LinkedIn, plattformen för professionellt nätverkande, uppdaterade för några dagar sedan sin policy gällande hatiskt och kränkande innehåll. Specifikt har man tagit bort formuleringar som inneburit ett tydligt ställningstagande mot trakasserier riktade mot transpersoner.
Läs hela inlägget: https://axbom.se/linkedin-raderar-tydligt-stod-for-transpersoner/

2025-07-18 08:04:03 UTC

I'm reading Catastrope Ethics by Travis Rieder. This passage drives home why we need to question how politicians talk so promisingly about negative-emissions tech that sucks CO2 from the air:
"The other category of negative-emissions technology avoids at least the land-use problem and is popular among the more techno-optimistic, as it does something that sounds borderline magical: it sucks CO, directly from the air and returns it to the ground. There's only one problem, which is that it doesn't yet exist in a scalable form.
What scientists have come up with so far are proof-of-concept models that show initial promise but that have very real drawbacks.
Consider, for instance, Orca, which is the largest carbon-capture plant in the world, located in Iceland just outside Reykjavik. This facility is instructive for many reasons. First, despite being the largest plant in the world, it will only capture about 4,000 tons of CO2 annually. And while that may sound like a large number, remember the scale that we're talking about: the all-time anthropogenic CO2 budget is 3.67 trillion tons. On average, humanity emits more than 1,250 tons of CO2 per second. What that means is that our largest plant to date, running around the clock for a year, will capture about 3 seconds' worth of humanity's emissions."

2025-07-08 14:44:00 UTC

- reply

Det fanns inte i min föreställningsvärld att statsministerns livvakter har telefoner som de får installera vilka appar de vill på.

2025-07-08 13:03:36 UTC

Jag har semester, men Säpo kan ju inte hålla sig i skinnet så det blev ett nyhetsbrev/inlägg ändå idag.
Dagens Nyheter visar i en granskning hur nuvarande och tidigare statsministrar kunnat kartläggas med hjälp av öppen information som publicerats av naiva livvakter i flera år. Detta inkluderar "mycket känsliga delar av flera skyddspersoners liv och vanor”.
Jag har använt Strava som ett varnande exempel i snart 10 år. Säpo har tydligen bott under en sten.
https://axbom.se/sapo-strava/

2025-05-12 08:16:37 UTC

I was reading a debate on LinkedIn where one side was arguing that "AI" image generators are more energy-efficient than hiring a photographer (or buying a stock photo).
One argument that made me do a painful face-palm was the notion that photographers need to eat, so they consume more energy.
The thing about humans is… wait for it… they ALWAYS need to eat. You can’t shift production to a computer and argue that it’s more efficient because it doesn’t eat. The human will keep eating(!) You are only adding energy-use.
These types of arguments illustrate a deep obliviousness towards human wellbeing. The argument only works if you expect the human to eat less when their job is taken away. Or die.
Their only data point for “AI" energy use was the point of generation, neatly excluding:
- all the energy used when scraping the internet for all the required datasets
- the energy used when building the model itself
- the energy used for required human workforce tasks, such as labeling (I bet those humans have to eat too)
- the energy used when building required data centers, both in-use and speculative investments
- the energy for hardware production and ongoing replacements
- the fact that the ease-of-use (and being “free”) encourages way, way more image generation than ever before…
Oh, and one more thing…
The image generator feeds on, and requires, photos and images taken and made by humans… who eat. Damn those humans.
Oh wait, the person doing the prompting… that person has to eat as well. It’s almost like it’s a human imperative.
Worth noting is also that none of the big "AI” companies are at all transparent about their energy consumption for this technology. Nor water consumption. Nor noise pollution. Nor e-waste. So there’s a ton of speculation.
I just wanted to make clear that I’m all for allowing humans to eat.

2025-05-05 10:04:08 UTC

- reply

It’s definitely an interesting point. One primary reason I switched from Mastodon to Akkoma was how bloated I felt the software was.

2025-05-05 07:32:39 UTC

When you’re running all your websites and apps on your own server, and traffic and space costs real money, you are way more careful with optimising images and files, and refrain from unnecessary uploads.
When people use “free" services, paying with their personal data and biometrics, they don’t seem to care.
This is one example of how free services are bad not just for privacy, but also for the environment – because all those costs are hidden.

2025-03-10 13:30:30 UTC

I know you know this, but it’s "sneak peek", not "sneak peak".
Unless you’re suddenly discovering a new summit on a mountain, that deviously was hidden behind a cloud.
Or, you know, if a climax takes you by surprise.

2025-03-05 18:36:29 UTC

- reply

Found this alternative in France now. Not as sleek-looking but is likely better in many ways… (on my iPad so haven’t been able to look at it in detail):
https://updown.io/

2025-03-04 20:19:53 UTC

- reply

Gjorde du allt på egen hand Anders?

2025-03-04 19:26:19 UTC

- reply

Ja något känns väldigt fel. Fullständiga backuper finns och ingår för alla VPS-kunder men min uppfattning är att de endast görs en gång i veckan. Jag har egna backuper på mina appar via Cloudron som ligger i storage hos Cloudflare, men jag vet inte hur det ser ut med mina möjligheter att läsa in dem ännu.
Fortsättning följer.

2025-03-03 07:00:14 UTC

- reply

Ja den där balansgången med enkelhet är svår. Ibland när jag pratar Mastodon föreslår jag bara att ladda ner appen Mona eller Tusky. Att särskilja protokoll, plattform och app är oerhört svårt för de flesta.
Att Signal går att ha på desktop först efter att man installerat den på telefonen stökar till det för en del också.

2025-03-03 06:50:59 UTC

- reply

För tydlighetens skull så var det inte jag som nämnde Signal i artikeln. 😉 Men när jag rekommenderar det (vilket jag gör i andra sammanhang) brukar jag i samma andetag även föreslå Matrix/Element.

2025-03-01 08:52:12 UTC

The biggest of pains for me, and difficult to admit, is not having moved my business email away from Google. I’m getting ready to do that within the next few months. I actually have my whole family’s email accounts in there as well, which has also added to the friction of moving.
Right now my biggest concern is identifying all the different other accounts and connections that may be dependent on that main google account, and if they contain anything of value. I’ve used this google account for so many years it’s bound to be attached to all sorts of things, and the ”plumbing” is not easy to grasp. As an example, the Feedburner service has been attached to a singular google account since google took it over (I no longer use it, but it’s a good example to understand the type of dependency I’m talking about, and that can result in not being able to access content).
Has anyone seen a writeup of what to watch out for when moving a domain name away from Google Workspace?
Realising I may have to do a writeup of my own in the end.

2025-02-23 21:35:29 UTC

My latest post/newsletter is now out. It’s a notable one where I share tips about software and services you can use to replace the big tech variants. I know people are looking.
https://axbom.com/big-trek/

2025-02-22 14:40:46 UTC

Very weird bug in Ghost... All of a sudden all the body text in the editor is grayed out. Everything still functions, and I can edit the text, but it's very hard to see(!)
What could be going on here?
* This is happening on all posts.
* It's happening in more than one browser.
* It's only happening for one of my Ghost installs (axbom.com), the others are still fine
Any ideas are welcome.
https://akmedia.axbom.com/5f82c85bff6e646ad46246e819bf007740def94f2b26460814fa340a4a2a5a13.png

2025-02-14 06:20:02 UTC

- reply

Intressant, Bridgy verkar inte alls plocka upp kommentarer/trådar? Hela tråden är mycket läsvärd men kräver inloggning på Bluesky:
https://bsky.app/profile/systerfrida.se/post/3li3eeqhz3c2h

2025-02-09 09:31:06 UTC

After he has so intently moved into writing about AI, usability superstar Nielsen has certainly doubled-down on making everything about IQ.
My post with quotes highlighting his obsession with IQ was updated today.
https://axbom.com/jakob-nielsen-quotes-about-iq/

2025-02-05 18:08:08 UTC

I want to see more pledges like this one by #Posthaven. We should be able to trust that our content can outlive us.
I’m sad for all the beautiful texts by great writers that have been lost because of how internet publishing tends to work.
Do check out posthaven.com
https://akmedia.axbom.com/7e9cd66ba0f0e0b854c11db46819820610d4781560f60dde11d83eb07cff9728.png

2025-01-30 07:12:08 UTC

- reply

Coding as a Sport: The Olympic Spirit in Programming 😁
https://codeweek.eu/blog/the-olympic-spirit-in-programming/

2025-01-30 07:09:12 UTC

- reply

Sant. Känns som någon borde ha skrivit om detta.

2025-01-30 06:51:52 UTC

- reply

Har aldrig tänkt på det så förut, men ja det är nog samma typ av driv som kickar in som när idrottare fortsätter springa fast de är slut.

2025-01-29 20:03:24 UTC

- reply

Couldn’t give up. Solved it. 🥳

2025-01-29 19:15:40 UTC

- reply

Too tired to figure it out right now 🥴

2025-01-29 19:08:59 UTC

I think I just managed to update my Akkoma instance.
I appear to have some settings wrong for the S3 configuration though… and that is likely why my images aren’t displaying. Like my avatar...

2024-11-30 07:41:25 UTC

I keep hearing…
"Everybody needs to start using AI!"
and…
"Ha! AI is everywhere. If you're using the internet you're almost certainly using AI."
Well, job done then.
Don't let the door hit you on the way out.

2024-11-22 10:03:44 UTC

What specifically happens if I close a Fediverse server and start a new Fediverse server with the same domain name, and I create the same handle as I had before.
I expect this to fail, but in what ways would it fail? Or would it actually work? I just don’t know. 😅 (besides losing all posts, but that always happens when moving servers on the Fediverse)

2024-10-02 12:45:45 UTC

You know how I avoid having a cookie banner on my website?
It's wild, but…
I DON'T SET ANY COOKIES!
My website is not perfect, but not using cookies is gosh so easy.
https://axbom.com/about/

2024-06-03 10:55:37 UTC

Heja Sverige. Vår statsminister går ut och gör gemensam sak med ett amerikanskt storföretag, och hyllar dem för att de gör en rekordinvestering i sina egna datacenter, bokar upp resurser på vårt elnät, gör åverkan på natur och miljö – och ger dem sannolikt rabatt på elpriset för besväret. Investeringen kan summa summarum bli en kostnad för civilsamhället.
Allt för något som kallas för AI, och som ingen har definierat vad det är för något. Ett marknadsföringsbegrepp som kan betyda allt och inget. Just det, vi får några timmar i deras webbutbildning – troligen utan lärare – om deras egen proprietära “AI”, en webbutbildning som sannolikt redan ligger färdigproducerad. Tack för det Microsoft.
Söker journalister som kan problematisera detta…
Det här var mitt första utbrott. Ska fördjupa mig i respektive ämne på bloggen sedan.
Just det… det råkar vara ett storföretag som kritiserats starkt på senaste tiden för sina bristande säkerhetsrutiner… bland annat av Trafikverket, och släpper en dator med en funktion (Recall) som är som att bjuda cyberkriminella på buffé.

2024-05-17 14:14:05 UTC

Instead of a traditional online forum I am considering starting a mailing list (think listserv). Some of my best best online conversations were on mailing lists in the late 90s. And e-mail is, after all, pretty cool.
Oh yeah, I’ve just been reading about how Slack is now ingesting your messages for their AI “training”.