Ex. this type of price estimation is quite naive - some models can require 2-3x the number of tokens to achieve the same level of intelligence. Artificial Analysis' own cost per task is a more fair estimation of cost.

[0] https://www.cerebras.ai/blog/cerebras-kimi-k2-Enterprise

Not sure on consumer/product use though

Even though this article is new, I'm not sure it has materially new ideas vs what some other labs have been claiming for years.

Enforcing safeguards that apply globally to good and bad actors alike so far as proved quite difficult - compute is essentially the only lever, and that's tenuous at best.

Unlike nuclear arms (where functionally there's a limit at MAD), so far we believe that with enough compute, energy, and algorithms, intelligence will continue to scale (and not only scale - but compound)

https://en.wikipedia.org/wiki/Uyghurs#Human_rights_abuses_ag...

I'm right there with you questioning Meta's value for society, but I do believe it has less opportunity for abuse/manipulation

2) What they already are good at they'll continue to improve at, because of domain knowledge & ecosystem

3) The reason they are ahead in the first place is labor cost & the QOL workers will tolerate (and intentional government investment)

4) The only way around this is for the US to invest into robotics (labor) & subsidize american manufacturing (ecosystem)

Post-training/fine-tuning is not trivial and having it as a service might make it more accessible.

[0] https://surgehq.ai/blog/training-on-complexconstraints

All of these things have historically been the job of engineers, because it helps other people contribute to the code.

Now it helps other people and other agents contribute to the code.

[0] https://www.notboring.co/p/the-electric-slide

Overall, I am in favor of this goal. I'm not sure this is the protocol I'd choose to accomplish it, but it's the one people hear about, and the one they're using.

I'm excited for when Github starts letting me check in the chain-of-thought that produced a line of code, and git blame it like I can with commits.

Engineering has moved up another layer of abstraction (just like we moved past managing buffers & writing machine code)

Or maybe it will be the more established open source model where the code is free but the maintainers offer hosting/some default product

It's not a big leap to apply that model to a company and its customers, where the company builds a well-abstracted, easily extensible base that 1) Customers can easily extend/customize for their workflows 2) Customers can self-host or run fully isolated, much easier (probably not quite there yet, but is a possible world)

These are more important than ever, because we don't have the crutch of "Teammate x wrote this and they are intimately familiar with it" which previously let us paper over bad abstractions and messy code.

This is felt more viscerally today because some people (especially at smaller/newer companies) have never had to work this way, and because AI gives us more opportunity to ignore it

Like it or not, the most important part of our jobs is now reviewing code, not writing it. And "shelfed" ideas will now look like unmerged PRs instead of unwritten code