I am so exhausted with thinking about AI, with talking about AI, with having to jump ship from one product to another because of AI, with having to explain that no, I don't have an {Instagram, Bluesky, whatever} account because of AI, and so forth.
-
That said, I'm also resigned to that this shit won't go away as long as there's so *so* much money being thrown into making AI inevitable and unavoidable. And that as long as this shit is here, it needs talked about and opposed.
I'm just exhausted, but not out of the fight.
@xgranade I also don't think it's just the money, I think part of the extreme push for making AI unavoidable is because they have exhausted easily extracted training sources (scraping the internet) and now need to reach for fracking (free training from your interactions using the AI).
Flaki (@flaki@flaki.social)
@devsimsek@universeodon.com > Human-generated data is irreplaceable. The “internet is running out of training data” problem just got mathematically formalized. Yeah I think the AI con mob has realized this already (but of course not saying the quiet part out loud). With Satya whining about people calling it slop and the AI industry trying to force it down everyone's throats no matter the cost (e.g. Copilot) I think they realize that there is only so much internet and historical content they can use to train their models - now they want *you* to help train it for them. Prompt Claude to spit out some code, ask Copilot for a PR review, and _interact_ with it, pointing out where it was stupid, confirming when it did a good job, by virtue of interacting with an AI model you are improving it with this exact, essential human input.
Mastodon (flaki.social)
-
That said, I'm also resigned to that this shit won't go away as long as there's so *so* much money being thrown into making AI inevitable and unavoidable. And that as long as this shit is here, it needs talked about and opposed.
I'm just exhausted, but not out of the fight.
@xgranade one of the points I'm actually midway through writing up right now is that the post that I am writing is *itself* a form of harm. there's like a variant of the "cross of iron" speech here in terms of the amount of adversarial communication we are being forced to engage in
-
@xgranade Also watching various prominent figures in the free/open source software space dismiss the ethical concerns over convenience/usefulness/whatever... which is *exactly counter to the point of free software as a social-political movement*.
@miss_rodent @xgranade this this this
-
@xgranade Also watching various prominent figures in the free/open source software space dismiss the ethical concerns over convenience/usefulness/whatever... which is *exactly counter to the point of free software as a social-political movement*.
@miss_rodent Yes! Maybe a hardline social-political stance is too far in the other direction at least for some, but it would be nice if some people would at least stop and consider *why* they are doing what they are doing and in the way they are doing it.
Because there's hardly a shortage of proprietary software in the world.
-
R relay@relay.mycrowd.ca shared this topic
-
@xgranade one of the points I'm actually midway through writing up right now is that the post that I am writing is *itself* a form of harm. there's like a variant of the "cross of iron" speech here in terms of the amount of adversarial communication we are being forced to engage in
@glyph I mean, there's a very good argument that AI is a wedge issue.
(As a side note, approximately 1700% of "this is a distraction" comments can be made more accurate by being replaced with "this is a wedge.")
-
That said, I'm also resigned to that this shit won't go away as long as there's so *so* much money being thrown into making AI inevitable and unavoidable. And that as long as this shit is here, it needs talked about and opposed.
I'm just exhausted, but not out of the fight.
@xgranade meanwhile every org that has embraced genai for code gen can no longer maintain any 9s of availability.
-
@miss_rodent Yes! Maybe a hardline social-political stance is too far in the other direction at least for some, but it would be nice if some people would at least stop and consider *why* they are doing what they are doing and in the way they are doing it.
Because there's hardly a shortage of proprietary software in the world.
@mkj @miss_rodent The failure mode of not thinking about ethics in open source movements is becoming unpaid volunteer scab labor for giant corporations.
-
That said, I'm also resigned to that this shit won't go away as long as there's so *so* much money being thrown into making AI inevitable and unavoidable. And that as long as this shit is here, it needs talked about and opposed.
I'm just exhausted, but not out of the fight.
@xgranade yup the key to all those "but AI is here to stay" arguments.
It won't go away as long as its receiving the big incentives from military contractors or government.
It is dependant on tons of money, aggressive data center spread and exploitation of people and resources.
-
I am so exhausted with thinking about AI, with talking about AI, with having to jump ship from one product to another because of AI, with having to explain that no, I don't have an {Instagram, Bluesky, whatever} account because of AI, and so forth.
I'm so tired of having to extend good faith to people uncritically regurgitating talking points from the worst techbros on the planet, of having to be polite about "but it works!" claims, of being asked to "set aside" the ethics.
@xgranade THANK YOU. Solidarity. So tired.
-
I am so exhausted with thinking about AI, with talking about AI, with having to jump ship from one product to another because of AI, with having to explain that no, I don't have an {Instagram, Bluesky, whatever} account because of AI, and so forth.
I'm so tired of having to extend good faith to people uncritically regurgitating talking points from the worst techbros on the planet, of having to be polite about "but it works!" claims, of being asked to "set aside" the ethics.
@xgranade some guy literally pissed me off because he argued that "ai is helping the entrepreneurs everywhere so it's worth it" and that "being vegan is far more environmentally friendly than doubling down on chat gpt prompts"
Frankly i was really childish with my response but holy hell
-
@xgranade THANK YOU. Solidarity. So tired.
@StaceyCornelius @xgranade we have a friend who is sort of a guru type, went to India and studied etc. Not a huckster or an operator, genuinely lives his life helping people.
Lately he's become endless about his "bot pal" which he's writing a book about/with. It is by turns annoying and worrying.
-
@StaceyCornelius @xgranade we have a friend who is sort of a guru type, went to India and studied etc. Not a huckster or an operator, genuinely lives his life helping people.
Lately he's become endless about his "bot pal" which he's writing a book about/with. It is by turns annoying and worrying.
@jpaskaruk I've been having quiet conversations in an online business group I'm in. I discovered that focusing on dependency and deskilling has influenced someone who initially saw LLMs as a big help to their business.
Between that and climate concerns, they've become a lot more hesitant.
Being an old school craftsperson (with a Fine Arts degree) can have surprising advantages these days.
"Want to be a better painter? You paint. You don't hire someone to paint for you." -
I am so exhausted with thinking about AI, with talking about AI, with having to jump ship from one product to another because of AI, with having to explain that no, I don't have an {Instagram, Bluesky, whatever} account because of AI, and so forth.
I'm so tired of having to extend good faith to people uncritically regurgitating talking points from the worst techbros on the planet, of having to be polite about "but it works!" claims, of being asked to "set aside" the ethics.
@xgranade Me too, but after 30 years of doing exactly that in every other tech corner I guess I'm a bit numb to it.
Honestly, I'm just about af the mind to unplug entirely, but I'm not sure how I'd pay my bills these days without it.
-
@jpaskaruk I've been having quiet conversations in an online business group I'm in. I discovered that focusing on dependency and deskilling has influenced someone who initially saw LLMs as a big help to their business.
Between that and climate concerns, they've become a lot more hesitant.
Being an old school craftsperson (with a Fine Arts degree) can have surprising advantages these days.
"Want to be a better painter? You paint. You don't hire someone to paint for you."@StaceyCornelius @xgranade "Like taking a forklift to the gym to lift your weights for you" is one of the best analogies I've seen.
-
@StaceyCornelius @xgranade "Like taking a forklift to the gym to lift your weights for you" is one of the best analogies I've seen.
@jpaskaruk Oof. I like that.
-
I am so exhausted with thinking about AI, with talking about AI, with having to jump ship from one product to another because of AI, with having to explain that no, I don't have an {Instagram, Bluesky, whatever} account because of AI, and so forth.
I'm so tired of having to extend good faith to people uncritically regurgitating talking points from the worst techbros on the planet, of having to be polite about "but it works!" claims, of being asked to "set aside" the ethics.
@xgranade
> I don't have an {Instagram, Bluesky, whatever} account because of AIConnecting a few dots, I think some folks are attempting to slip AI into the backbone of the Fediverse:
- Scott Jenson, on Mastodon GmbH payroll, arguing for AI tolerance on Fedi
- Mastadon GmbH winning grant money to fund centralized denylists and centralized content labelling
- Nivenly being involved in the FIRES specification, FIRES being the spec Mastodon GmbH stated they are following
- Around the time Nivenly tried to make TBS the central denylist, folks (not sure if associated with Nivenly or not) were openly saying that human driven moderation was ultimately futile and that AI moderation was needed
- Talk of using "automation" (didn't say AI, but I'm pretty sure it's code for AI) for content labelling
Veering on conspiracy theory maybe. I don't think I'm far off. A bit more quiet and with coded language this time. Except from Jenson who's trying to normalize AI on Mastodon / Fedi.
I'm dubious.
-
@mkj @miss_rodent The failure mode of not thinking about ethics in open source movements is becoming unpaid volunteer scab labor for giant corporations.
@xgranade @mkj @miss_rodent
every FOSS contributor is either the last bastion of computer justice or Mark Zuckerberg's plumber
bonus points if they're playing both campaigns at once XD -
@xgranade
> I don't have an {Instagram, Bluesky, whatever} account because of AIConnecting a few dots, I think some folks are attempting to slip AI into the backbone of the Fediverse:
- Scott Jenson, on Mastodon GmbH payroll, arguing for AI tolerance on Fedi
- Mastadon GmbH winning grant money to fund centralized denylists and centralized content labelling
- Nivenly being involved in the FIRES specification, FIRES being the spec Mastodon GmbH stated they are following
- Around the time Nivenly tried to make TBS the central denylist, folks (not sure if associated with Nivenly or not) were openly saying that human driven moderation was ultimately futile and that AI moderation was needed
- Talk of using "automation" (didn't say AI, but I'm pretty sure it's code for AI) for content labelling
Veering on conspiracy theory maybe. I don't think I'm far off. A bit more quiet and with coded language this time. Except from Jenson who's trying to normalize AI on Mastodon / Fedi.
I'm dubious.
@xgranade
Nivenly became suspicious to me for supporting (funding?) an image genAI OSS project that became notorious for being used to generate CSAM. They eventually dropped it, but not before trying to double down on supporting it."We will build a reference implementation of a Automated Content Detection service, again as a new Fediverse Auxiliary Service Provider with an open protocol"
https://blog.joinmastodon.org/2026/04/sovereign-tech-agency-funding/"As a Data Provider annotating … with Labels enables both machines and humans to better understand the information being presented. For instance, a Data Consumer may choose to automatically apply recommendations with one label, …"
https://fires.fedimod.org/concepts/labels.html"The writing of the proposal initially outlining FediMod FIRES was funded by Nivenly Foundation"
https://fires.fedimod.org/"AI Horde has also been…making sure that our public resources on generative AI images do not facilitate the creation of CSAM…In this case we’ve had to put weeks of work…"
https://nivenly.org/blog/2023/08/14/a-deeper-dive-into-haidra/ -
@xgranade
Nivenly became suspicious to me for supporting (funding?) an image genAI OSS project that became notorious for being used to generate CSAM. They eventually dropped it, but not before trying to double down on supporting it."We will build a reference implementation of a Automated Content Detection service, again as a new Fediverse Auxiliary Service Provider with an open protocol"
https://blog.joinmastodon.org/2026/04/sovereign-tech-agency-funding/"As a Data Provider annotating … with Labels enables both machines and humans to better understand the information being presented. For instance, a Data Consumer may choose to automatically apply recommendations with one label, …"
https://fires.fedimod.org/concepts/labels.html"The writing of the proposal initially outlining FediMod FIRES was funded by Nivenly Foundation"
https://fires.fedimod.org/"AI Horde has also been…making sure that our public resources on generative AI images do not facilitate the creation of CSAM…In this case we’ve had to put weeks of work…"
https://nivenly.org/blog/2023/08/14/a-deeper-dive-into-haidra/"…we announced that Haidra has joined Nivenly as a member project."
"our decision is to follow the sentiment of the membership in the vote and not accept Haidra as a Nivenly project."
https://nivenly.org/blog/2024/04/15/haidra-decision/[I'm having difficulty finding sources of criticism of Haidra facilitating CSAM. It can be inferred from Haidra's own blog post having rushed (my opinion) CSAM detection and somewhat inferred from the Nivenly community vote to unsupport it.]
Scott Jenson pro-AI (while denying that characterization), appropriating Black experiences, and other problematic things 🧵
https://social.coop/@scottjenson/116352522288234148That isn't a source for every claim in my first reply, but I'll stop for now.
-
That said, I'm also resigned to that this shit won't go away as long as there's so *so* much money being thrown into making AI inevitable and unavoidable. And that as long as this shit is here, it needs talked about and opposed.
I'm just exhausted, but not out of the fight.
@xgranade
me too. all of this.
