Show more

Why does everything in this space come across as so damn sleazy?

Show thread

From @travisj:

1. genAI is theft
2. genAI devalues labor
3. genAI is inefficient
4. genAI is inaccurate
5. genAI is resource intensive
6. genAI learns all our bad habits
7. genAI gives power to the wrong people
8. genAI makes the rich richer and the poor poorer
9. genAI isolates
10. genAI disrespects open source licensing
11. genAI pollutes the web with junk and spam
12. genAI won't teach me anything

#zuzai

futurehorizondesign.net.au/pos

automating tasks displaces anger and aggression && prevents the real issues from ever reaching those who can address them ( from a French text on automation of public services shs.hal.science/file/index/doc )

automating tasks makes auditing the final result take longer & the human auditors lack the mental context and the intuitive understanding of trade-offs made & mistakes can cascade and degrade the entire ecosystem & incentive is strong not to see this: freakonometrics.hypotheses.org

#fuckgenAI

RE: circumstances.run/@davidgerard

I also picked 2024 as the year AI would have gone bust because that was one year after “pilot programs” started in mega corpos. You’d think one year of showing shit-all for ROI would have ended the experiment, but nooo

So:

- Leftist Lula wins in Brazil 🇧🇷
- Progressive student movement wins in Serbia 🇷🇸

Come on USA, I know you can kick the orange douche bag to the curb! 🇺🇸

#Eleicoes2026 : #Lula victory speech:

"We had to overcome foreign interference through #digitalplatforms, which we will have to regulate.

We cannot have monstrously huge corporations earning millions & billions, paying no taxes to this country & trying to steer this country's election results."

I understand the point of this [formerly serious newspaper that has devolved into fluff] profile of #MarissaMayer (former CEO of Yahoo!) may be rage bait, so I am not including a link to it.

But ouf, reading these words (which have been widely circulating online) has been painful, especially as a mom.

You can't think of 3 qualities that make your child special, so you ask an AI Assistant? NO, NO, NO, NO. UNACCEPTABLE.

I imagine Marielle reading this and my heart breaks for her.

#NoAI

@simondev @RosannaSibora @_elena Think talks and workshops like this. @sfscon seems to understand. 🫶

sfscon.it/workshops/the-future

And btw I find diversity focused activities the peak of cringe, but until we are more than 10% of a sector that we all depend on for our digital sovereignty, it remains necessary. Kudos to the @fsfe for organizing this workshop.

Wow, this is pretty astounding. 48% of practicing Christians say they would “trust AI with their spiritual growth.” 44% of Millennials say that AI’s spiritual guidance is “as trustworthy as a pastor’s”.

It shouldn’t have surprised me that a group of religious people who are desperately seeking simple, clear, unambiguous answers to difficult questions are falling for the machine that gaslights people.

This study (n=1514) was done by Barna Group, an Evangelical Christian polling group.

#AI #christianity #religion

“AI is Becoming a Spiritual Authority, Even Among Practicing Christians”

barna.com/research/christians-

RE: mstdn.social/@DemocracyMatters

I understand and share the joy of dunking on Trump, but please don’t do it this way:

- IQ is largely a nonsense metric

- Given its context and history, using IQ as an insult is playing with eugenicist fire

- Purporting to quantify “intelligence” at all is wildly problematic (we don’t even have a good definition of the thing we’re measuring here, folks)

- Purporting to quantify intelligence of someone at a •distance• is even wildly-er problematic

- Measuring human intelligence with an LLM is so off the problematic charts I don’t even know where to start

Find a better insult. It’s like shooting fish in a barrel with that guy. (“Like Saddam Hussein opened a Cheesecake Factory” was a pretty damn good one.) And if you (rightly) want to puncture his wild lies about his superintelligence, focus on his •lying• and not his IQ. Please and thank you.

I saw this on Reddit today and it made me sad. Sad that LLM propaganda has polluted the minds of young people in such a universal way. This person is apparently a student who has to complete some IoT project, but has no experience. He says: "I had to use AI codes to help with this". Sad that he feels that way and doubly sad that he doesn't see his mistake in trusting that the statistical guessing machine could magically write his project for him. Now he asked for humans to help him😡

New Mexico ain’t fuckin’ around. I am disappointed that the court saw fit to allow Meta to continue preying on seniors and adults, further poisoning our public discourse, but at least they’re holding Meta to account for a portion of their crimes.

nmdoj.gov/press-release/court-

Ex-Manny Yekutiel spokesman now works at Anthropic, another S.F. political staffer at an AI giant

Failed supervisor candidate, former assistant district attorney, and more have ended up Anthropic and OpenAI....

missionlocal.org/2026/10/conno

prototyping is the slow part that makes the final build a matter of assembly.

quick prototyping just pushes the yard full of rakes into the final build process

Show thread
heeyyyy so jsyk the #french government, due to the currently ongoing mass protests, attempted to pressure ISPs into blocking access to social networks.

we only know this because ISPs refused and the official meeting letter was leaked to the press.

this is some proper authoritarian bullshit. human rights country my ass. spread the word, the world has to know what's happening in france.

story according to @blast_info
https://mamot.fr/@blast_info/117497423566287403
#france #frpol #fr
Show more
Librem Social

Librem Social is an opt-in public network. Messages are shared under Creative Commons BY-SA 4.0 license terms. Policy.

Stay safe. Please abide by our code of conduct.

(Source code)

image/svg+xml Librem Chat image/svg+xml