It is nice that there are 6 more questions after that, but honestly - it's already over after the first.

Mood

Many critical AI resources are written for teachers, rather than students. I wrote this for undergrad students dealing with “AI literacy” in the humanities, hoping to raise more fundamental questions about what it means to write. People have added it to syllabi, so maybe you’ll find it useful, too.

Read this thread. I don't know what to say, other than we can't live like this anymore.

AS A: Employee I WANT TO: Not use this software but I have to SO THAT: I don't get fired ---------------- The enterprise SaaS user story

you don't need to add caveats agreeing that the head of state abducted is a bad guy before opposing abduction of a head of state by military force. If the rule was any head of government is fair game then the entire world would be a series of decapitation strikes and show trials in foreign courts.

This all seems pretty unbecoming of a FIFA Peace Prize winner

I've lost count of the number of times just this week that I have accidentally clicked an "AI" button in a service or software, which means a bunch of services are counting me as a regular active "AI" user despite me being about as hardcore "no generative AI, nowhere" software guy as they come.

i’ve been interested in darts for the whole of 20 minutes, and i already do not like Littler

a nice piece on the Resonant Computing Manifesto gameboat.bearblog.dev/the-resonant...

Someone on Mastodon sent me this brilliant cartoon that sums it up nicely

Remember when people said it was crazy to think that we were in an AI bubble? When they would actively mock you for doubting the glory of the LLM era?

Plagiarism is a crime. Say it with me kids, Plagiarism is a crime, all LLM’s are plagiarism. 😑 Tech promised for years that the final product would not rely on scraped LLM to run their AI…

There's a thing to weave into African Speculative Fiction -- "When the AI bubble burst, it was from the detritus of an exploitative industry that Africa built their foundation in the coming age of technology. The PCs of a thousand failed AI startups abandoned in their centers spawned a revolution."

...

"video game mechanics?" Actually, they're called developers.

Over ten years ago I pitched a piece about how all these glowing profiles were being done about 3d printing companies "disrupting" and "democratizing" prosthetics and how that was simply not true. I was told it was too boring/negative and nobody wanted it. www.marketplace.org/episode/2025...

You would be surprised, if you hadn't worked in the field, just how dogshit a lot of these tech people are at actually doing tech stuff.

absolutely harrowing account of a tech worker named michael from nairobi kenya forced to act as an ai sex bot. how is this not classified as sexual abuse bordering on trafficking???? data-workers.org/michael/

not the first time this happens blacksky.community/profile/did:...

I was at something recently where a guy who worked in tech told a grad student that all the problems with facial recognition were fixed now and have been for a long time. So the student shouldn’t talk about it in their diss.

But the people who read articles aloud on YT and TikTok get brand deals and loving praise from their commenters on their “journalism” the people who do the underlying reporting work for companies that increasingly don’t exist and get laid off and are attacked as corrupt/fake news or whatever

the word layoffs doesn't even appear in this article It's just all analysts chirping away about synergies media coverage of itself and its endless consolidation doesn't think labor, consumers, or creatives are actually relevant to anything

the thing about online queer discourse is unless you dig into things it's hard to tell if someone is speaking from a place of nuance and decades of lived experience or whether you're reading the untested absolutes of the most 15-year-old alive

No. No it can’t. At least not accurately. And we should think LONG & hard about what it means to “diagnose” something you have no intention of bringing resource to treat. Through eye scans. That could be collected without someone’s express consent. This version is a recipe for weaponized med tech.

Part of the problem is the only things that need the data centers are the 100B+ weight LLMs. But super-massive LLMs are, shockingly enough, not very useful because they can never be used when correctness matters or they might receive untrusted input & do a bad thing.

You can solve nearly all confusion about geographical distance when talking to Europeans by simply adopting the Poland as a unit of measurement. Example: "I haven't been to any of the conventions on the west coast, because they are at least 3 Polands away from where I live."

The real question is not "Why is Bluesky so left-coded," the real question is "Why can't the right exist in a social media environment without algorithmic assistance, where people can ignore them"

Immigrating is so hard. To tell an 8 year old that she hasn't effectively demonstrated that she can't be cared for by relatives in Haiti instead of her parents in the UK is just bare malevolence.

My "Not involved in financial fraud" T-shirt has people asking a lot of questions already answered by my shirt.

Why is the largest company on the stock market arguing with random points in this memo? why is it written in the most defensive tone possible? This feels like 3 accountants and 4 lawyers tried to stop Jensen Huang from posting it and got overruled

It is wild to me that the largest company on the stock market, one with such incredible prestige and a huge PR team got dog-walked here. Why would you EVER give this ANY oxygen? Now everybody is asking whether NVIDIA is like Enron! This didn't help at all!

🚑🕊️ The Red Cross stands for “neutrality” and “impartiality”. Our investigation shows Russia’s branch trains children in military skills, works with Kremlin linked groups, and operates in occupied Ukraine while still receiving millions from a system funded by EU governments. Full story below:

imagine that. imagine that. imagine that! imagine that. did anybody say anything about this once or twice or thrice, hmm? did anybody talk about, this? no? hmm?

oh you GOD DAMN THINK?

*sydette just shrugs at this point * We need to have a convo about research and access and approaches because people spent a lot of 2020- 2024 shoving folks asking for different methodologies because we don’t trust META and they said We didn’t “understand”

This is also why the platform companies, despite very obviously being publishers, are so desperate to avoid being labeled as publishers. It's all about evading accountability for things any normal company would be burned to the ground for.

do you have any idea how much you have to piss somebody off to make them change their gauge

If a product caused harm, or death, to a single person who was using it correctly, the person responsible for that product should be in a state of distress for what they did. Doesn't matter about the scale of the product. If they didn't say how to use it correctly, they are no less liable.

y'know, if tech companies weren't so fired up to call every single thing "AI" it would be a lot easier to distinguish "ML spam filter that is well-sandboxed and tuned on your emails" from "giant context and IP-eroding foundation model that makes people paranoid and unhappy"

I've seen a particular drop in quality of the code produced by my coworkers leaning on LLMs. Code running is absolutely not the benchmark for code quality. Being able to read and understand the code so that you can fix problems when they arise is. Any jackass can trial and error into running code.

happy Jair Bolsonaro got arrested after failing to cut off his ankle monitor day

This is a systemic problem I will be writing more about elsewhere, but there's a tendency to see medicine and medical research as a solution to defective people, rather than as the prerequisite for all other basic freedoms. The catch is that we then question whether defectives deserve the fix.

Recent updates have turned on features that I had already, explicitly turned off. That just undermines principles 4,5, and 8 of Mozilla’s principles, and you should review your internal procedures to prevent that from happening again. www.mozilla.org/en-US/about/...

PayPal has rolled out a new "feature" that prompts you to leave feedback about an artist that you've commissioned via invoice. Whatever you do, DO NOT fill this out! To quote Admiral Ackbar, "IT'S A TRAP!!!" It is meant to misguide you. (continue...)

Then let people choose to opt-in, rather than forcing them to learn how to opt out. Especially when the technology being foisted is as useless and deceptive as A.I.

I did skip it entirely. It changed the settings against my will. You keep cramming it in our faces. We hate it. Knock it off! AI sucks and is hugely wasteful of both time and resources. It‘s an insult to your users

What PISSES ME OFF relentlessly about all this is the gaslighting: tech industry: "AI will make everything faster/easier" Reality: AI creates an incredible amount of work for me trying to uninstall, escape, move to whole new platforms/tech to get away from it