Comment by wordofx
4 days ago
Why wouldn’t you?
The only reason you wouldn’t is because you get upset with Elon. It’s not a bad model. It’s leagues ahead of anything meta has managed to produce.
4 days ago
Why wouldn’t you?
The only reason you wouldn’t is because you get upset with Elon. It’s not a bad model. It’s leagues ahead of anything meta has managed to produce.
Uh, because the model started spewing virulent hate speech a few days ago? What normal software does this?
Not the model itself, the X bot. Its obvious that this has happened due to them tweaking the bot, you could never get it to write anything like this a couple of weeks ago.
Can you trust the model when the people releasing it are using it in this way? Can you trust that they won't be training models to behave in the way that they are prompting the existing models to behave?
An acute memory will remember this happening with basically every chatbot trained on text scraped from the internet, before they had to explicitly program them to avoid doing that.
Any LLM trained appropriately. Tokens in, tokens out.
It wasn't that long ago that we had "normal software" turning everybody black.
This is just how AI works, we humanize it so it's prone to controversy.
There have been a few recent instances where Grok has been tuned to spew out white supremacist dreck that should be political anathema--most notably the "but let's talk about white genocide" phase a few months ago and more recently spewing out Nazi antisemitism. Now granted, those were probably caused more by the specific prompts being used than the underlying model, but if the owner is willing to twist its output to evince a particular political bias, what trust do you have that he isn't doing so to the actual training data?
xAI has over 1000 employees. If he was polluting the model we would know about.
If?
https://x.com/elonmusk/status/1936493967320953090
He seems pretty open about it.
Who was responsible for the "kill the Boer" dreck? Were they disciplined? Did they get fired? Why don't we know that?
1 reply →
I think it's far more likely there are a tiny handful of mid-tier unhinged sycophants among those 1000 employees who think that pleasing Elon means polluting the model to make Grok an unhinged sycophant, because that's what an unhinged sycophant would think to do.
Elon explicitly ordering this? Press X to doubt.
Why should these topics be outright banned?
Are you asking in good faith why non-sequiturs that stoke racism shouldn't be injected into unrelated twitter threads?
Even related twitter threads, do you want interracial hatred to be increased?
Here's why it should be banned, because it leads to this shit:
https://en.wikipedia.org/wiki/The_Holocaust
https://en.wikipedia.org/wiki/Slavery
https://en.wikipedia.org/wiki/Rwandan_genocide
https://en.wikipedia.org/wiki/Apartheid
https://en.wikipedia.org/wiki/Armenian_genocide
https://en.wikipedia.org/wiki/Gaza_genocide
https://en.wikipedia.org/wiki/Unit_731
How can people be so fucking stupid that they want to be coy about recreating the most shameful atrocities in human history? Teenage idiots a few years removed without any understanding of the world beyond their nose and brains turned to putrid rot.
The direction that right wing reactionaries are taking the world in could not possibly be more disgusting and pathetic.
32 replies →
> Why wouldn’t you?
Because its poisoning the air in Tennessee?
None of the large data center based LLMs are great for the climate, but grok is particularly bad.