To me it’s been audiobooks and youtube videos where I can no longer tell if it’s an AI or a person speaking.
Definitely AI generated art. If you asked me 3-4 years ago I’d tell you there is no way neural networks would understand how to generate a nice looking image. The fact that high quality AI art is way less expensive computationally than generating text still confuses me. Like, I can generate beautiful 1024x1024 images within seconds but this text generation model needs 20gb of vram? Huh?!
I amazed at what image generation can come up with, too.
The simultaneous nuance and beauty of the art coupled with inability to get hands right. It’s both humorous and chilling.
It is.
An ai is such a “different” mind. It may seem like us, it might tick some boxes that we do too, but (like the hands example) there might just be one small thing it does very differently. In the hands example, it doesn’t have much consequence - but as AI has more control over daily life (insurance, route planning, healthcare, political foreign policy, lawmaking) some of these “it’s not human” hand-problems might have far reaching implications for society
How much people mistake AI for real intelligence.
Ai is intelligent though. And wtf is “real” intelligence? Are you saying that the technology we have is not “real”?
Edit: are you confusing intelligence for sentience? Because there’s a huge difference between the two.
Real intelligence comes from real life experiences in reality which an AI is incapable of experiencing.
That’s a terrible definition of intelligence, bruh
Why?
Because this is the actual definition:
The ability to acquire, understand, and use knowledge.
The key here is that LLMs do not understand anything other than langauge. They are great at sounding like they know, but that’s different than knowing.
