Battle your pickles.

"Claude, I am really tired. My kids do not see me enough, my wife is threatening to leave me. Users are frustrated, my manager is pushing for some new features, there are some security pen tests coming up. Could you help me out here?"

Can someone who is into the nomenclature explain to me if Pete Hegseth is a centaur or a cyborg with respect to AI?

I think about this Tony Benn speech much more than I used to

One of my unfounded conspiracy theories about generative AI is that to fake the "intelligence" they simply packed in a very high unpredictability factor. Rogue actions feel like a choice, a decision of intellect, while predictability does not; predictability is already served by non-AI software.

2 years of seeing my friends and colleagues in the climate movement doing this has radicalised me

i’m tired of every paper (even critical ones) on societal impact of AI starting by both-siding the “benefits” and risks/harms of AI in academia, i would like us to arrive at a collective reckoning that we don’t need to play both sides. it's totally legit to clearly state just the harms. period

The so-called AI era feels like the part of a Despicable Me movie where all the minions have turned purple.

I see my friend @malk-zameth.bsky.social is doing a talk on "out of the tar pit" and would love to do an "into the tar pit" co-talk at some point.

"Stack superfluous"

Nobody wants to know what you and your chatbot talked about

Har sagt dette mange ganger, men: hvis du driver med NOE SOM HELST KREATIVT og bruker AI til å promotere det er jeg ikke interessert. Du kan spille i verdens kuleste band. Er plakaten eller platecoveret AI kommer jeg ikke til å sjekke dere ut.

From a discussion in RISF based on the old IBM adage, an updated version for the modern era:

Software developers be like what’s the purpose of architecture if I still have to design?

Mark Twain and Benjamin Franklin uttered fewer self-help gems than one might think.

In my slide decks, I've started including quotes of what AI leaders believe to be true, and helps break through the notion that they are some sort of genius magicians. This context matters, if you are to decide about using and trusting (!) the products they make.

Have said it before and will say it again: No one will ever convince me that a machine or artificial system deserves more rights and dignity than the human with the least. If your project is convincing me of the worth of the former, better out your resources into improving that of the latter.

Teach a man to fish and you'll feed him for a lifetime, sell a man a subscription for fish and he'll grow dependent upon you for survival and you can jack up the price over time.

Code is computational microplastics.

Oh wow looks like I'm going to Antwerp in June.

I find this whole "AI agent wrote a hit piece" thing really troubling for reasons that have nothing to do with bots getting "mad" and "deciding" to take down open source contributors. theshamblog.com/an-ai-agent-... Why do we keep wanting to give AI so much agency that it lets humans off the hook??

"From whatever to non sequitur: a guide to AI-generated talk titles"

Good Grief!

#JavaZone is built by the community. Maybe this year, that includes you. 📍 NOVA Spektrum, Lillestrøm 🗓️ September 2–3 Free access to sessions, great people, real event experience. Sign up here: 2026.javazone.no/vol...

Sam Altman er en niding

Vibe koding er for nidinger.

What a time to be alive for people who have been pushing the systems perspective in software.

Poor NewCrafts conference, imagine having "raising the bar" as your motto during the current bar-lowering avalanche.

Jeg ville tenkt at folk som driver med arkiv kanskje burde være de siste til å bruke systemer som finner opp fakta.

Nasjonalarkivet har forkastet hele forsiden av nettsiden sin til fordel for et KI-tekstfelt? Har dette kanskje gått litt for langt nå?

I'm pretty sure I know. They aren't. We're not going to really be able to make a lot of progress in dealing with the implications of this tech unless and until we get rid of all this "woo-woo" talk about LLMs. Anthropic pushing this line is PR, unserious.

I continue to be bewildered that AI boosters don't get that no one considers it a feature that their most clueless coworkers can generate mountains of text objects for other people to deal with, or that companies have another way to avoid their users.

"Skamvett" is a useful Norwegian word that translates roughly to "sense of shame", but sense as in wisdom (good sense) rather than feeling. It denotes something that seems to be in short supply these days.

As they say: "Planning is essential. Plans are useless" AI-tools help you skip the planning, and go straight to generating the plans! 🥳🧠

Developers until yesterday: Lines of Code are a terrible way of measuring productivity, not to mention value delivered. Developers today: The LoComatic 3000 has just been released! Wheee! Code go brrrr!

Are you anxious or are you simply correctly assessing the numerous threats

The problem is not so much the artificial intelligence in itself as the natural stupidity it brings out in some people. And the dehumanising systems they propose to build around it.

One thing generative AI has definitely taught us is that there are a significant number of people who can't tell the difference between something average and something excellent - a LinkedIn post by me.

I struggle to articulate the depth of ignorance, vacuousness, arrogance, contempt and devaluation of human beings that is imbued in statements like "AI agents are your new co-workers".

"I never had high expectations of this industry, but it still managed to disappoint me." <- This

I had no idea people spent that much time booking hotels.

By the end of this abstract you'll be tired of generated prose.

Senior developers, 2026: "Remember StyleCop, the noisy syntactic linter that had 3 rules that did no active harm and 197 that did? I have set up an AI agent to act as StyleCop for me now."

2026: Et tu Brute for all the Brutes.

Human in the loop -> Human in the slop.

My cat just threw a 422 Unprocessable Entity on the table.

Bob: Did you just ask the chatbot to tell me to go fuck myself? Ed: I believe I did, Bob.

A short blog post where I propose a radical new heuristic to be applied before replying to things online: ask yourself if the original poster could already know. einarwh.no/blog/2026/02...

All O'Reilly book covers should feature pictures of depressed and/or distressed owls until technologies drastically improve their relationship with existence.

iOS update, yippee.