This is the fediverse, so this is probably going to be extremely unpopular, but yea here goes.

Not saying you have to love it, or trust it blindly. Skepticism is healthy. But there’s a difference between “I use it carefully and check everything” and “I refuse to touch it.”

Historically, refusing new tools has never stopped them. Coopers didn’t stop the steel drum by refsuing to make them. They just ended up out of work while the industry moved on.

The realistic threat isn’t AI replacing you. It’s a colleague who uses AI well doing your job faster. And the skill of using it properly (knowing when it’s wrong, veriyfing output, knowing what not to use it for) takes time to build. Starting late means competing with people who’ve had years of practice.

You can have legit concenrs about ethics, energy, or jobs and still learn the tool. Opting out doesn’t fix those problems. It just removes you from the people who get a say in how it’s used.

PS. Ethically trained models exist and you can run models locally.

you are viewing a single comment's thread
view the rest of the comments
[–] 0 points 9 hours ago

It's unlikely people will be priced out of it. Not in any meaningful way. That logic shows a misunderstanding or ignorance of the current landscape.

One of the biggest unspoken problems is the good enough effect.

Local and mod scale models that can run efficiently and meet or exceed the needed utility level for most small to medium applications is basically here right now.

Which undercuts the big companies entire business model to some degree.

At current pace, a skilled developer can use the existing models to do real work. While your avg person with no skill can only really do very small scope things or generate slop.

But if those models keep at pace then larger and larger workloads keep falling past the good enough line. Each time that happens the entire business model of the big ai companies becomes more and more shaky.

Open weight models by their design undercut the very value and service the big companies are selling.

The same thing happened in game hosting on a smaller scale. You had dedicated small companies early for huge premiums.

But then as the software got better, more efficient faster, you stopped needing specialty technology a lot of compute. And you could start hosting it on your own computer or a small server. At that point it became more efficient to sell cheaper and in bulk than it was to sell a specialty good.

All those early movers in the market died out basically overnight and then all that was left were the small bulk sellers and self-hosters.

If openweight models keep at their pace, there is a legitimate chance that it's going to be cheaper to buy a bunch of old server racks and cheap older gpus in a generation or two to run models at a fraction of the cost instead of paying these big companies.

The only thing these big companies could then sell is management in other services around it. Which is still a fraction of the value and profit of what they're doing now which still is going to collapse their entire business model.

Lot of people seem to be acting like gpus are just never going to get any faster as well. And a faster GPU into more efficient GPU means that models go farther and farther.

A lot of the value right now comes from the simple fact that really only one generation maybe too. Arguably worth of GPUs are even viable for this sort of work.

And two or three generations of gpus. We're going to see speeds and efficiency. Make what we're doing now seem kind of quaint.

Yeah the models will be bigger than and more bulky and need that more compute. But again the only thing that matters is the good enough line.

If you could run anthropics mythos class frontier b******* on $1,000 server that you're leaving in your little small business is closet. Then even if they never got any better at all from what they are right now.

We have passed the good enough line.

The entire industry is self-cannibalizing and and basically living in a pretend world as if technology is never going to advance. In everyone seems to be in the same delusion. The haters most of all seem to be forgetting technology advances and improves with time.

And that means that these companies can't keep doing what they're doing because what they're doing requires technology to stop advancing and it's not going to stop.

  • source
  • parent