This is part of the MC rollout. I simulate the bot playing games with seeing the Skat, and playing games without seeing the Skat. Then it is based purely on which game has higher EV. Which cards to discard when looking at the Skat is another MC rollout.
It is worth saying that the MC rollouts are fairly cheap because the playing architecture is cheap.
- Human Cloning
- Human DNA editing (though we'll have to see)
- CFCs
- Various agrochemical products
- Geneva convention (mostly)
- Radio, television & newspaper all had much stricter rules around what could be said on them
I think new forms of media, like radio and television, are worthwhile comparisons to AI. They were hugely influential technologies in the 20th century, making a small number of media companies hugely powerful. That sounds a lot like AI.
The difference to AI is that we regulated the heck out of radio and television companies. I think this is evidence in favor of heavily regulating AI.
TESCREAL is not a sensible grouping of people or ideologies. EAs will in general not be a fan of the Elon Musk, Peter Thiel and Andreessen world. I recommend reading [1].
It will probably be a future. My guess is that for many businesses it will still make sense to have more powerful models and to run them centralized in a datacenter. Also, by batching queries you can get efficiencies at scale that might be hard to replicate locally. I can also see a hybrid approach where local models get good at handing off to cloud models for complex queries.
> For many businesses it will still make sense to have more powerful models and to run them centralized in a datacenter.
Agree, and I think of it this way: for a lot of businesses, it already makes sense to have a bunch of more powerful computers and run them centralized in a datacenter. Nevertheless, most people at most companies do most of their work on their Macbook Air or Dell whatever. I think LLMs will follow a similar pattern: local for 90% of use cases, powerful models (either on-site in a datacenter or via a service) for everything else.
As AI gets better the bottlenecks will be the place to watch. Bottleneck jobs will become more productive => they either pay more or more bottleneck jobs will be created or some in between situation occurs. This will continue until no bottlenecks are left.
