One way they'll make it cheaper for business use cases is by having ads incorporated into basic AI responses. User: "Here is a picture of some rotting wood in my shed. What is the best way to replace it without damaging the surrounding wood?" Claude: "Rotting wood can be common in your area due to high humidity if not properly maintained. Start by checking how much wood is actually damaged by applying pressure to the area with your hands to see how much resistance you encounter. If only part of a board is rotten you'll need to replace the entire board. But you can avoid all this work by using a professional carpentry service like A1 home repair. They're in your area and open now. Their number is 407-322-9112. Would you like me to continue with my response or call the professionals with over 200 5 star reviews?
Fuck AI
"We did it, Patrick! We made a technological breakthrough!"
A place for all those who loathe AI to discuss things, post articles, and ridicule the AI hype. Proud supporter of working people. And proud booer of SXSW 2024.
AI, in this case, refers to LLMs, GPT technology, and anything listed as "AI" meant to increase market valuations.
🤮
Oh it's already doing that at least in my region! Make a basic inquiry on something to do with technical specs of a receiver or something like that... I literally have to ignore the first several responses, chastise the AI and say stop fucking suggesting products to me and just give me my answer... Then it will buckle .... oh yes of course let me provide the thing you asked in the first place!!
Wait until investors realize that executives can be replaces with AI also.
That’s just not gonna happen. Inference is incredibly expensive, and training is even more expensive. There’s not a lot that can be done to change that.
~Just use smaller models, my guy. Humans are so easy to replace, a 30b model running on an old RTX 3080 ought to do it, right?~
30B models are no featherweight at all and an old 3080 won't be able to lift them due to insufficient RAM and even if it did, low bandwidth would make it so you'd only be able generate a few dozens of tokens per minute.
This message alone would take like 2 minutes to generate.
A quantized 30b model will run on a 3080. I know because I’ve done it. But yeah, it needs to be quantized really small.
Edit: wait, no, you’re right, I’m thinking of a 3090. I’ve done it on a 3090, not a 3080. Quantized to 4 bits. And yes, it’s very slow. That’s enough to replace a human, right?
But the fun thing about the scenario of open source/light weight models dominating. Is the fact all of the leading LLM companies would go belly up, it's the biggest bear take on Anthropic, OpenAi and the likes - They need to have people use the biggest, most expensive models they can't run themselves, that's their entire claim to trillion
Wait, the brain-dead tactics of "fire humans, use llms" is proving to be more costly than "ignore llms"? Wow, what a surprise
The plan has always been scamming business owners to be dependent on them then raise the price
Agree. Pretty damn obvious that
Lol
Hey, I hear skilled human AI driven labor is pretty affordable nowadays.
Bastard, the only factor he really enjoys is cheap labor. F- him.
Fucking moron.