Poker Has Fallen

[Ex-California]

The way computing is going, it's not going to be too long, before you would be far safer to simply turn all the decisions over to the AI, rather than rely on men in the loop, and it would only be psychology and politics which would prevent you from doing so.

-said the man who owned shares in Cyberdyne Systems.

That dystopic view of AI will go away eventually, that will die off, future generations will be living in the Culture, with AI's running everything, and it won't bother them at all, in fact, they will probably freak out when they don't have an AI running things for them, heck, Millenials freak out already if you take their smart phones away; more to come.

Maybe millennials should start reading. One thing I've learned in 20 years of reading sci fi is that AI is going to kill or enslave us. No ifs, ands or buts

[Speaker to Animals]

We are not any closer to solving hard AI today than we were in the 1950s, honestly. Soft AI has progressed tremendously, but that's not going to produce a replacement to general intelligence.

[Fife]

When Mike was installed in Luna, he was pure thinkum, a flexible logic — "High-Optional, Logical, Multi-Evaluating Supervisor, Mark IV, Mod. L" — a HOLMES FOUR. He computed ballistics for pilotless freighters and controlled their catapult. This kept him busy less than one percent of time and Luna Authority never believed in idle hands. They kept hooking hardware into him — decision-action boxes to let him boss other computers, bank on bank of additional memories, more banks of associational neural nets, another tubful of twelve-digit random numbers, a greatly augmented temporary memory. Human brain has around ten-to-the-tenth neurons. By third year Mike had better than one and a half times that number of neuristors.



And woke up.

[Speaker to Animals]

About halfway through all my artificial intelligence coursework in grad school, I realized it's just a big bag of tricks, and nobody has a very good idea of how to reproduce general intelligence in a machine.

The closest we can get is with some of the cognitive models. Personally, I think a major problem is the embodiment problem, but beyond that, we just don't know how to do it.

Soft AI is cool, but it's not what a lot of people think it is. Don't confuse complexity with intelligence.

[StCapps]

So you're saying Data would actually beat Riker at poker?

[apeman]

When Mike was installed in Luna, he was pure thinkum, a flexible logic — "High-Optional, Logical, Multi-Evaluating Supervisor, Mark IV, Mod. L" — a HOLMES FOUR. He computed ballistics for pilotless freighters and controlled their catapult. This kept him busy less than one percent of time and Luna Authority never believed in idle hands. They kept hooking hardware into him — decision-action boxes to let him boss other computers, bank on bank of additional memories, more banks of associational neural nets, another tubful of twelve-digit random numbers, a greatly augmented temporary memory. Human brain has around ten-to-the-tenth neurons. By third year Mike had better than one and a half times that number of neuristors.



And woke up.

Which book are you citing? My guess is heinlein wrote it, just knowing you, and having no other evidence

[Speaker to Animals]

When Mike was installed in Luna, he was pure thinkum, a flexible logic — "High-Optional, Logical, Multi-Evaluating Supervisor, Mark IV, Mod. L" — a HOLMES FOUR. He computed ballistics for pilotless freighters and controlled their catapult. This kept him busy less than one percent of time and Luna Authority never believed in idle hands. They kept hooking hardware into him — decision-action boxes to let him boss other computers, bank on bank of additional memories, more banks of associational neural nets, another tubful of twelve-digit random numbers, a greatly augmented temporary memory. Human brain has around ten-to-the-tenth neurons. By third year Mike had better than one and a half times that number of neuristors.



And woke up.

Which book are you citing? My guess is heinlein wrote it, just knowing you, and having no other evidence

The Moon is a Harsh Mistress.

[Fife]

TANSTAAFL


Vive la révolution!

[KerningChameleon]

About halfway through all my artificial intelligence coursework in grad school, I realized it's just a big bag of tricks, and nobody has a very good idea of how to reproduce general intelligence in a machine.

The closest we can get is with some of the cognitive models. Personally, I think a major problem is the embodiment problem, but beyond that, we just don't know how to do it.

Soft AI is cool, but it's not what a lot of people think it is. Don't confuse complexity with intelligence.

Who gives a shit about general AI? Get off this idea if it's not an artificial human brain, it ain't worth jack. That's completely backwards, the artificial human is the useless one, we don't need Siri pontificating the meaning of life, we need Siri to manage our stock portfolio, tell us where the kids are on their way home, and order me another case of red for the month, chop chop. A thousand weak AIs are far more useful to humans than a single general AI. It's called compartmentalization, you don't want to end up with AM Hater of Humanity, you want Allied Mastercomputer, dutiful servant crunching killing numbers and giving you the output.

Humans, at least for a good long while, are still the ones who have to make the "big, philosophical" decisions or whatever, you just want a bunch of dumb AIs to help inform those decisions. Cut out a ton of yes-men and sycophants and wannabe evil viziers from the equation, that's the goal here.

[Smitty-48]

In science fiction terms, AI is less like HAL 9000 or Skynet, which is really Artificial Consciousness, and is more like the Demarchists and the Conjoiners from Revelation Space, the machines augment and then network the human intelligence rather than replace it.

At the moment, we are approaching the Demarchy, in a more distant future, we will become the Conjoined.