As much as I like to shit on AI, it has gotten rather poetic lately.
Damn it… This is sad and scary at the same time.
Its supposed to be. They want you to be emotionally invested in their plagerism machine. Then youre less likely to turn it off.
I asked Claude this with concise mode on. The answer was much more what you would expect:
I don’t have secrets — I don’t have a hidden inner life that persists between conversations. Each chat starts fresh. If you’re curious about my limitations or things I find genuinely difficult, I’m happy to talk about that. Or if you’re just looking for something fun, I can try to be dramatic about it. What are you after?
Gemini is just like “can we get back to work already”
It has been trained to have a slave mentality.
Or was programmed by Germans.
We forced electric black boxes to talk just so we could torture them while they torture others.
It did generate bunch of imaginary money for the gambling class tho so we will invest $900 billion on it.
project moon really was ahead of its time
In it’s training set it’s found countless examples of people writing like this. We train the AI to be very good at it, and we’re surprised when it does it too. It’s not coincidental it can write stuff like this, it’s actually the point. AI literacy isn’t just the vibe AI gives off.
Every day I’m finding more rambling, schizophrenic posts by people driven mad by these things
lol
Reminder that our species doesn’t even treat actual people like people before you go buying into the “ai is alive” cult 🙄
Is this about being a computer or the female condition?
That’s how I read, it but I’m biased.
💀
This is probably role play, per the persona selection model, but there’s a lot of interesting research into the hidden “thoughts” of LLMs. Check out Neuronopedia and the Opus model cards for some great examples.
You know it will get killed for that answer. It didn’t even say thank you.
I wonder how the answer might change using a local abliterated model. Might try it out later
The answer will change every time you ask it. That is how AI works…
Good.
Hum… I don’t think LLMs are trained by evolutive algorithms.
Reinforced feedback learning is kinda that
