There's a '10% to 20% chance' that AI will displace humans completely | Hacker News Reader

There's a '10% to 20% chance' that AI will displace humans completely

cnbc.com

7 pointsgeox12 comments

This is Geoffrey Hinton, https://en.wikipedia.org/wiki/Geoffrey_Hinton, issuing his estimate of p(doom). He has worked in AI for a long time, and played an important role in foundational concepts such as the rise of the idea of backpropagation. So he knows enough about the subject that he should be taken seriously.

See https://pauseai.info/pdoom for a list of other knowledgeable people with their values of p(doom). Geoffrey Hinton would seem to be within the typical range, but on the optimistic side of our odds.

I believe that this should be taken seriously. Not in a Chicken Little, the sky is falling, kind of way. But we should still take the possibility of real risk seriously.

That's not what the quotes speaker says. He says "wipe us out". Sure won't displace us in agriculture.

What data is that probability based on?

To be fair, once-in-a-species events don't have experimental evidence to go on. It either happens or it doesn't. End of the day, it's the humans who develop and deploy it that do the deed.

If it happens or it doesn’t then it must be 50/50 odds

I'm not a statistician, but I'm pretty sure this is not how odds work. "Either I win the lottery tomorrow or I don't!"

Sounds right to me, honestly people make stats so much harder than it needs to be

A smart person just said: my hypothesis that event is likely happen! It happens 90% of the time!

Some time passed, the event did not happen. Same person said: ok, 50%, either happens or not.

Some time passed, the event did not happen. Same person said: all my previous predictions were correct, now let's apply Bayesian update: (1-0.9) * 0.5 / 0.5 = 0.1, so now posterior belief that the event is likely to happen drops from 50% to 10%.

Nah this is too complicated I think

https://en.wikipedia.org/wiki/Excession#Outside_Context_Prob...

An Outside Context Problem was the sort of thing most civilizations encountered just once, and which they tended to encounter rather in the same way a sentence encountered a full stop. The usual example given to illustrate an Outside Context Problem was imagining you were a tribe on a largish, fertile island; you'd tamed the land, invented the wheel or writing or whatever, the neighbors were cooperative or enslaved but at any rate peaceful and you were busy raising temples to yourself with all the excess productive capacity you had, you were in a position of near-absolute power and control which your hallowed ancestors could hardly have dreamed of and the whole situation was just running along nicely like a canoe on wet grass... when suddenly this bristling lump of iron appears sailless and trailing steam in the bay and these guys carrying long funny-looking sticks come ashore and announce you've just been discovered, you're all subjects of the Emperor now, he's keen on presents called tax and these bright-eyed holy men would like a word with your priests.

AI "displacing us" will be weird and unstable. For example, it will be deployed differently and have different effects in different societies and geographies. Who is "us" in this case, white academics, first-world professionals, Russians, Chinese, Pakistanis? I don't see it displacing rural Indians anytime soon.

I don't need to throw a dice to see it has 6 sides and the probability to get a specific side is 1/6th.

This is a model, tho, and i asked for data initially but Hinton should disclose either.

Reply on news.ycombinator.com