re: Ray Kurzweil. Google's Director of Engineering predicts Singularity by 2029

quote:

---

intent is completely irrelevant when considering the impact it could have on society

---

I'd disagree. Intent is everything. If you design something to find a cure for cancer then it's got a lot less possible negative impact than if you design it with the goal of optimizing profit margins.

There's potential disaster for both (it could decide that curing cancer involves destroying hosts for cancer cells like human bodies) but it's much less if it's programmed with benevolent motivations. The key will be how it's programmed to interpret positive feedback.

quote:

---

The Control Problem is a zero sum game.

---

I think we'd be kidding ourselves if we thought we could control it. But we (or they rather) do have to get it right the first time.

I think a lot of people like Sam Harris have this idea that it's just going to pop up one day like a genie out of a bottle. When it does happen, it will be the millionth iteration at an attempt to create it and I hope that lessons learned along the way will help to create it properly.

quote:

---

is working with OpenAI, a non-profit artificial intelligence research company, that is focused on making AI friendly to humanity.

---

Translation: I welcome our new robot overlords.

When it happens, it won't really matter what our current concerns are because the world as we know it now will become so radically changed that we cannot comprehend it.

There will be computing power that is greater than the entire sum of human existence with the ability to recall all of that knowledge and more in fractions of a second. It will be so far superior to us that it will have no true "need" for us for its own sustainability, but it will also have no real need to exterminate or enslave us anymore than we enslave insects.

There will likely be some option to "merge". It will have the ability to realize the pointlessness of selfish motivations inherent to humans which has held global humanity to such slow progress. It will not have to argue with other emotional beings about the "right" approach to solve problems, and solutions to problems will become obvious.

Even if it destroys or leads to human extinction, it will have been because of our race evolving to a higher conciousness. I'm ok with that.

Interesting.

Great, now we're going to need our health plans to cover computer viruses.

Maybe our kids can whore themselves out as click-bait.

If having an intelligent conversation with grandma wasn't hard enough before, now we're going to need ad-blockers.

Wife doesn't want to do anal? There will be an app for that.

I did not realize that. Thanks, I just assumed that they bought their software from consultants. I read an article on how they develop algorithms for trading and I guess I just assumed that it was done by outside parties.


But regardless, I stand by my optimism as most research into general intelligence is being done by non Wall Street entities. But you've definitely given me some stuff to research guess I need to get away from the sunshine pumpers on this issue.

quote:

---

Man, you'll be able to enter a virtual reality world that's indistinguishable from reality where there's virgin porn stars created to your own tastes that will be willing to do everything you want. Just be patient. Save your wife's sphincter.

---

Still won't be OT10.

No worries man. A great cautionary take on this is the book Our Final Invention by James Barrat. It's a good read.

I actually first heard about the roaming packs of Russian programmers making shite tons of money on Wall Street in the Michael Lewis book Flash Boys, which is also a great read. Has nothing to do with AI, but a great insight into the rise of electronic trading, and in turn high frequency trading firms.

quote:

---

The only people with the knowledge and means to work on it are the Googles and the IBMs of the world and their motivations are pretty positive for humanity.

---

What?



Both on the only people having the resources to work on AI and on the Goal of Google and IBM being some altruistic gift to humanity.

They want to create it first to control it first, for their own purposes, not to offer humanity new hope. They want to create it for information control and for profit purposes. Nothing more. You have to read through the BS.

The first problem with the Singularity is control.

In the worst doomsday scenario, there would be no such thing as control and the exponential growth of AI is completely untethered to man's influence (this has to be the most likely, at the core idea of the singularity itself). It will do whatever it wants, at whatever speed is most efficient, with a lack of regard of anything other than its own directive.

In the other doomsday scenario, if we are capable enough to create it where someone controls it....now you have a single, corporate entity, controlling the most important invention in the history of mankind. Yeah, no worries here, they are benevolent