Create accountLogin

Recent searches

No recent searches

Search options

Only available when logged in.
eupolicy.social is one of the many independent Mastodon servers you can use to participate in the fediverse.
This Mastodon server is a friendly and respectful discussion space for people working in areas related to EU policy. When you request to create an account, please tell us something about you.

Administered by:

Server stats:

213
active users

eupolicy.social: AboutProfiles directoryPrivacy policy

Mastodon: AboutGet the appKeyboard shortcutsView source codev4.3.8

#grok

7 posts7 participants2 posts today
Public
Mark Carrigan

When LLMs engage in a flamewar on social media

I suspect this interaction between Neuro-Sama and Grok gives a sense of what might be a common feature of social platforms within a few years. Grok enters at 7 minutes if you get bored of Neuro-Sama’s thread:

https://www.youtube.com/watch?v=B2FhGyNWDXQ

What happens if most people encounter LLMs through their social media personas? How will that drive public perception of models and consumer uptake of them? For example I could imagine ChatGPT, Gemini and Claude coming across as deeply boring in these situations (though I suspect Claude Opus might get subtly irritable being cast in these situations) suggesting that we might see optimisation for public performance.

en.wikipedia.orgNeuro-sama - Wikipedia
#content#Grok#LLMs
Replied in thread
Public *
RS, Author, Novelist, Prosaist

@TheZeldaZone

If #AI were good … companies would [not be forcing] us to use it.

Very good point. The objective is to reduce labor costs, that is, to fire all knowledge and tech workers. It's not that it is better, it is that it is cheaper. Of course, this completely misses the point that someone has to generate the knowledge, the #writing, the images, as well as the raw #science data upon which to train the AIs, which requires people with experience doing the work.

This will not end well.

#BoostingIsSharing

#theComingLaborApocalypse#genAI#LLM
Public
Beardy Star Stuff

I'm finding it hard to come up with words to discuss this. The wealthiest oligarch in the world, obviously also a murderous psychopath willing to poison entire communities. The string of obsceneties that I want to type would only begin to express the rage. How are these people allowed to become such monsters? Why do we tolerate their violence, their cruelty? Why do we allow them to exist?

FUCK YOU FUCKING FUCK.

#ElonMusk #Oligarch #Grok #AI #ClimateEmergency

youtube.com/watch?v=3VJT2JeDCy

YouTubeWe Went to the Town Elon Musk Is PoisoningBy More Perfect Union
Public
Ketan Joshi

Elon Musk is choking a black community with air pollution to run a racist chatbot on a right-wing social media site. And the next data centre planned to run "Grok" will nearly double the number of fossil fuelled turbines.

Far from 'saving' the climate, it really looks like fossil-fuelled data centres are running climate-denying racist text generators for right-wing billionaires.

I found a new dataset that lays it all out pretty nicely: as bad as the generative AI data centre panic-build is, Musk is the worst of a bad bunch by a good margin.

Enjoy:

Links ->

The dataset: epoch.ai/data/ai-supercomputer

Southern Environment Law Centre update: selc.org/news/resistance-again

Thermal images of the gas turbines: theguardian.com/technology/202

More Perfect Union short documentary on the town Musk is poisoning: youtube.com/watch?v=3VJT2JeDCy

Musk's chatbot is racist: reuters.com/business/musks-xai

Once the site is grid connected, it'll raise power prices and increase grid-scale fossil fuel use: floodlightnews.org/power-for-d nytimes.com/2025/05/16/busines

A nice case study of a *defeated* 3.5 gigawatt data centre fossil fuel power station combo selc.org/news/a-rural-virginia

#ElonMusk#XAI#Grok
Public
Miguel Afonso Caetano

RT @mario_gug
Of course, based on Elon Musk’s legitimate interest:
"third-party “collaborators” can train AI models on the site’s data unless users #opt_out. X also feeds user data into its AI model, #Grok, for training."

theverge.com/news/680626/x-ai-

An illustration of the X logo
The Verge · X’s new policy prevents companies from using posts to ‘fine-tune or train’ AI modelsBy Emma Roth
Public
PrivacyDigest

#ElonMusk Claimed the #Art This Man Painstakingly Created Was Generated by #Grok

Over the weekend, Elon #Musk shared Grok altered #photographs of people walking through the interior of instruments and implied that his #AI system had created the beautiful and surreal images. But the underlying photos are the work of artist #CharlesBrooks , who wasn’t credited
#copyright

404media.co/elon-musk-claimed-

404 Media · Elon Musk Claimed the Art This Man Painstakingly Created Was Generated by GrokCharles Brooks knows that people will share and alter his work. He just wants credit for his photos when it happens.
Public
Flipboard Art & Photo Desk

Last weekend, Elon Musk shared some images on his X platform that he claimed had been created by the Grok AI system. They were actually just edits of original work by Australian artist Charles Brooks, a former concert cellist who is known for photographing the insides of instruments. An X user named Eric Jiang had run some of these through Grok, adding people to the original photos so they look like buildings. Musk retweeted the post with the comment "Generate images with @Grok." @404mediaco's Matthew Gault spoke with Brooks. “I feel like, if he’s plugging his own AI software, he has a duty of care to make sure that what he’s posting is actually attributed correctly and is properly his,” he said. “But ‘duty of care’ and Musk are not words that seem to go together well recently.”

flip.it/OQhrJ3

404 Media · Elon Musk Claimed the Art This Man Painstakingly Created Was Generated by GrokCharles Brooks knows that people will share and alter his work. He just wants credit for his photos when it happens.
#Art#AIArt#GenerativeAI
ExploreLive feeds
About