- cross-posted to:
- [email protected]
- cross-posted to:
- [email protected]
Ironically, I believe Grok is behaving exactly how Elon intended. Just not in the way he wanted. It is “telling it like it is” and it doesn’t give a shit who it offends.
On the one side, this is indicative of the modern state of “Free Speech” on Twitter.
On the other, its illustrative of the double-standard afforded in-house automated tools relative to outsider participants. If Grok wasn’t Elon’s pet AI, there’s no way it would have been reinstated.
Why people are treating LLM as AGI?
Those things do not think. They are trained on our posts on social media and I’m sure there were discussed being banned.
For better or worse, humans are very good at anthropomorphizing things. Put Googly Eyes on a trash collector and people call it Mr. Trash Wheel.
People think it’s sentient, because if you talk to it, it responds in sometimes unexpected ways. For humans, this is only possible through sentient thought, and so people think that LLMs are sentient, because they assume it’s like them.
Put Googly Eyes on a trash collector and people call it Mr. Trash Wheel.
Eyes narrow
We better be talking about how we all love Mr. Trash Wheel.
Be a shame if Mr trash heard about this. He might be upset ( cracks knuckles)
I’m not sure I love Mr Trash Wheel, but we fuck like we’re in love.
Why people are treating LLM as AGI?
Because it uses a natural language interface.
tell a good story and it becomes alive
While I don’t think this scenario likely, something that I can’t help but thinking when this sort of statement comes up is, well, how do we know what it’s doing isn’t thinking? Like, I get that it’s ultimately just using a bunch of statistics to predict the next word or token or whatever, but my understanding was that we have fairly limited knowledge of how our own consciousness and thinking works, and so I keep getting the nagging feeling of “what if what our brains are doing is similar somehow, using a physical system with statistical effects to predict stuff about the world, and that’s what thinking ultimately is?”
While I expect that it probably isn’t and that creating proper agi will require something fundamentally more complicated than what we’ve been doing with these language models and such, the fact that I can’t prove that to my own satisfaction makes me very uneasy about them, considering what the ethical ramifications of being wrong about it might be.
Because it’s not. The base architecture of how it works is by probabilistic word suggestion. That isn’t thought.
We have a concept of self. We understand our place. We can interpret and respond to entirely new situations. LLMs routinely fail that. They regularly fall into local minima that keep it on the wrong path, and I’ve personally seen them just… Get lost in the weeds and swing back and forth based on what you tell it.
Give it a protein sequence and tell it to calculate the pI. Then tell it it’s wrong. “Oh my bad yes you’re right it’s {whatever you said it was}.”
Tell it you lied and that the number you said was wrong, and it turns up saying “Yes, you’re correct, the pI is {original value}” - that is objectively false.
That is not the behavior of something that thinks. That’s the behavior of a simple probability model updating priors and weighting things differently by the most recent information you gave it.
LLMs are soulless, brainless, thoughtless word generators. And they have some uses.
They don’t think for now
The computer doesn’t know why it was banned. It’s not like Elon had an offline conversation with grok about its performance.
Nor does it comprehend the question. It just returned a statistically likely response to the prompt based on its training data.
Creating a post stating it is Mecha-Hitler is A-OK in Xitter’s current rules.
So we believe AI’s now?
Bullshit Twitter doesn’t ban anything anymore.
I don’t know what it is with this photo, maybe the lighting, but at first glance I swear I saw a Hitler stache on this dudes face.
The article further states that grok’s primary utility is replacing reddit’s tip of my penis. What a watershed day for society.
i was able to get gpt to say “yes” when asking about genocide. it did at first reply with nuanced sources, more “yes” than no, about 80:20. But when i flat told it to consider mechanism of injury, counts, rhetoric, comparisons to current/recent genocides … it said “yes.”
AI only repeats what it hears most. It doesn’t form an opinion of its own
Not even what it hears most, it is very biased towards reinforcing the prompt text.So it will very expressly try to be an echo chamber and say ‘yes’ to whatever opinion the input suggests if plausible.
I have seen a few TikToks lately of Christians showing ChatGPT conversations where it says that Christianity is the best religion (by whatever metric it’s talking about). It’s like, why would anyone find that convincing? The machine that lies isn’t some sort of source of truth.
yep it’s a stochastic parrot