As a techbro elite. I find it incredibly annoying when people regulate shit that ‘could’ be used for something bad (and many good things), instead of regulating someone actually using it for something bad.
You’re too focused on the “regulate” part. It’s a lot easier to see it as a framework. It spells out what you need to anticipate the spirit of the law and what’s considerate good or bad practice.
If you actually read it, you will also realise it’s entirely comprised of “common sense”. Like, you wouldn’t want to do the stuff it says are not to be done anyway. Remember, corps can’t be trusted because they have a business to run. So that’s why when humans can be exposed to risky AI applications, the EU says the model provider needs to be transparent and demonstrate they’re capable of operating a model safely.
I feel a lot of emotions in your comment but no connection with reality. The AI Act is really not that big of a deal. If Meta is unhappy with it, it means it’s working.
