Canalys: Companies limit genAI use due to unclear costs (www.computerworld.com)
from tonytins@pawb.social to technology@lemmy.world on 16 Jun 21:36
https://pawb.social/post/26503203

#technology

threaded - newest

etchinghillside@reddthat.com on 16 Jun 21:39 next collapse

Monetary or other?

tonytins@pawb.social on 16 Jun 21:40 collapse

Monetary.

Glitchvid@lemmy.world on 16 Jun 22:33 collapse

Not too surprising, it takes a 100kW AI rack to accomplish a fraction of what I can wrt writing code, and I can run on tacos and diet coke.

hendrik@palaver.p3x.de on 16 Jun 23:56 next collapse

Don't companies also limit the number of computers, staplers and pens in the office? ...It somehow has to be worth it. And they have different contracts available, you can set a limit with most APIs, bookkeeping can look up how much they paid... I think working somewhat efficiently is a normal part of doing business.

tonytins@pawb.social on 17 Jun 00:14 collapse

Normally, companies put the burden onto to us, which allows them to quickly scale, but because AI costs a fortune per-prompt, they got kinda shot themselves in a foot.

hendrik@palaver.p3x.de on 17 Jun 00:29 collapse

I mean sometimes we get that with other things as well. Like wasting cloud storage permanently. Or printing full color images on the expensive printer. Sometimes there are expensive supplies which shouldn't be wasted. Idk, kind of depends on the job. If you're a bartender or pizza chef, you'd also track the expensive ingredients and not serve an arbitrary amount. I guess AI is about the same.

blargle@sh.itjust.works on 17 Jun 01:42 collapse

Just wait until all the venture capital OpenAi raised on a valuation that assumes they will singlehandedly achieve the singularity in 2027, replace all human workers by 2028, and convert 75% of the Earth’s crust to paperclips by 2030 runs out, they can’t operate at a loss anymore, and have to raises prices to a point where they’re actually making a profit.