Yeah, I guess the debate is which is the lesser evil. I didn’t make the original comment but I think this is what they were getting at.

If you aren’t allowed to freely use data for training without a license, then the fear is that only large companies will own enough works or be able to afford licenses to train models.

Can be good experience to move away from home for a bit…

Since it’s processed on device they don’t (necessarily) need to transmit and store your conversations in some central location. I guess theoretically this could be done in a secure way.

Yes. It’s clickbait because it doesn’t tell you the name of the game.

Tape

*digital license agreements

I think you need a forest, a pulp mill, and a paper mill. The forest might be the hardest part in a lot of places…

We can save shingles if we stop vaccinating now!

Glaciers are there year round.

I don’t think you are allowed to sell patent infringing products, even if you didn’t make them. The manufacturers may not be in the US so going after them could be more complex. I believe patents restrict rights to make, use, or sell.

As long as you have enough monitors for all the spreadsheets and wiki pages you need to consult!

No Candy Crush? Non-starter.

I just use a second phone to record the first one.

Imagine hiring a taxi without having your lawyer review the terms and conditions, lol

I’m not sure that’s true for pedestrians? Going over the hood on a sedan might be better…

Battery failure https://spacenews.com/directv-fears-explosion-risk-from-satellite-with-damaged-battery/

But you have to select if it was human or not, right? So if you can’t tell, then you’d expect 50%. That’s different than “I can tell, and I know this is a human” but you are wrong… Now that we know the bots are so good, I’m not sure how people will decide how to answer these tests. They’re going to encounter something that seems human-like and then essentially try to guess based on minor clues… So there will be inherent randomness. If something was a really crappy bot then it wouldn’t ever fool anyone and the result would be 0%.

Isn’t most of the AI training work in the world done on Linux using Nvidia GPUs (in the cloud)? I guess it’s a different use case…