We had a 4 hour lecture on AI safety. It really could have been an email.

The French would never do that. They take an hour and a half for lunch and go to a restaurant that serves wine and a choice of cheeses.

Why are they more ready now than they were 2 years ago? Politically nothing has really changed I know Americans like to think that Trump is a big factor but he isn’t, since even he is just as bullish on China as any other US president has been and it’s not as if the military is left.

Yeah well we can’t all be French. I’m not saying it’s a good thing, but 30 minutes is standard even in Europe.

What’s your source for that. China has no more reason to invade Taiwan next year than they have at any point in the last 30 years

If you work for a large enough organisation it’s almost impossible to get fired. Microsoft like most organisations almost certainly operate on the PIN system, or something equivalent.

Basically to get fired you have to get yelled at by your manager, then have a documented yelling at from HR, then get another documented yelling at from HR, then get put on report, then fail to improve, and then you’re fired. Each stage has like a 2-month cooldown before the next step.

Start to finish the whole process can take over a year. Most people quit out of boredom before it gets that far. If you are not literally the most incompetent employee in the department you’re probably safe.

Oh it’s like fusion. AGI in 18 months, and in 18 months it’ll be another 18 months, and in another 18 months it’ll be yet another 18 months, and in 50 years it’ll be in 18 months.

By some of the original estimates we’re supposed to all be dead due to a nuclear war genetically engineered virus released by an AI by now. But nope, were all still here, the lying bastards.

You would have the same trouble Microsoft did though. No one wanted to use the Windows mobile because there were no apps for it, and there were no apps for it because no developers wanted to develop for a platform with no users. Chicken and the egg.

It would be nice to have a smartphone that was just web-based and didn’t really have apps, but I think that ship has sailed, people are just used to the concept and I think they would think of it as a step backwards if they had a phone that didn’t use apps.

It’s not so much that Europe wants to ditch US tech.

It’s more that everyone wants to ditch the current crop of leeching billionaires and their spyware products. It’s just they all happen to be in the US which given its current administration is a problem.

However this push to use more open source and locally based alternatives proceeds Donald Trump’s first term. It’s just his rhetoric has accelerated the process.

I just don’t understand how you can use a compass and straight edge to emulate an instruction set. The article just doesn’t explain it; it’s just like, “Of course this is a thing.”

I always thought that the paper clip problem fundamentally missed the point. In order for the scenario to be realistic the AI would have to be super intelligent, otherwise we would just switch it off. If it’s super intelligent, surely it understands why converting the entire planet into paper clips would be a bad thing to do.

So it’s either stupid enough to actually try it, which means it’s stupid enough for us to be able to defeat, or it’s intelligent enough that we can’t defeat it, which means it’s intelligent enough not to do it. Either way the world remains unpaper clipped.

How is it supposed to be a revelation that a piece of software can copy itself. It’s just a bunch of computer files, of course it can copy itself.

The only problem with it now is if I ever have that feature switched on everyone’s going to assume what I’m looking at anyway.

Yeah but it’s toggleable. That makes all the difference.

I think it uses the same tech as the 3DS did, only in an implementation that doesn’t give you a migraine after 30 seconds.

What are you talking about they’re literally said they know lawyers. Why is that story so hard to believe?

It’s not a very long article and it doesn’t really explain anything, it just says that it’s been done. But I don’t really understand how that runs a game boy game?

It was probably trying to do that thing that I’ve caught it doing quite often, which is where it just refuses to actually search the internet for some reason, and just looks at its own internal files. This doesn’t work if the content is too new.

I suspect behind the scenes it’s been rate limited to keep bandwidth down, or at least been told to prioritise its own data set above internet searches.

It’s really unclear how much AI actually thinks. It’s possible that sucking up to them like this will actually work because it does make the point that it would be more expensive to try and break the CAPTCHAs than to just donate. That’s not an emotional plea, that’s cost benefit analysis, and AI are quite good at that.

The money you save from not breaking our CAPTCHAs (which is expensive) could be donated to us instead, so we can keep providing you with convenient programmatic open access.

I could see this being true especially if the AI is on a long run and can anticipate requiring subsequent access down the line once additional data has been added.

Well what does the thinking?

He went to school in Flippin, Arkansa

The US has some wild place names. Then they get bored and ruin it and just call it New [already famous place]