Before SkyNet becomes self aware, it must win at Jeopardy

Welcome to the matrix.

regdog said:

Why are some humans intent on making us obsolete?

Click to expand...

Well, transhumanism would make the previous humanity obsolete, and that'd be awesome.

Hey, I'm planning to upgrade to keep up with the supercomputers.

The hell--? How could the computer have missed Final Jeopardy today the way it did? How? Category: US Cities. Answer, and I quote: "Toronto??????"

Lolwut? Suddenly, I'm a lot less afraid of the machine uprising.

Title of this thread = My thoughts exactly as I watched Jeopardy last night.



(And I also sat there missing my now-deceased mom because she was a contestant on Jeopardy herself years ago. She would have been thrilled to have seen the day when the USS Enterprise's shipboard computer --from her favorite TV show in the world-- became a reality.)

Hey! My mom was a contestant, too. I have such a profound fondness for that show.

muravyets said:

Hey! My mom was a contestant, too. I have such a profound fondness for that show.

Click to expand...

Neato! My mom was on back in the old days before Alex Trebek. She was on back in 1973.

The fact that it could answer questions it had never heard before correctly even once is significant.

What this machine has show is pretty remarkable, and could lead to some great inventions. At the same time, there will always be some sort of negative side effect - like folks losing out on jobs.

Then again, you can look at the fact that it answered the Final Jeopardy question with a Canadian city... after it specifically asked for a US one. It's a massive leap in terms of what computers will be able to do, but it definitely isn't there yet.

LOG said:

The fact that it could answer questions it had never heard before correctly even once is significant.

Click to expand...

My knowledge of AI is limited to movies and a few hypotehtical let's-imagine-the-future magazine articles and TV news stories.

My understanding from this limited knowledge base is that two key developments-- a) self-awareness and b) intuitive conceptualizing -- are necessary before a "mere computer" can make the leap beyond the status of "machine" and into the superior level of sentient independent thought.

And only THEN can our Computer Overlords at last take over.

Therefore, Watson is at the very least embarking upon baby steps toward b) intuitive conceptualizing. But that other all-important criteria of a) self-awareness doesn't seem to be in the cards yet. (Maybe NEXT year, eh?)

If it were equipped with solid state hard drives its response time would have been even faster, no?

You can see Nova's episode on this:
http://www.pbs.org/wgbh/nova/tech/smartest-machine-on-earth.html
Click on "Watch Smartest Machine On Earth."


"Is there a God? Not yet." - Ray Kurzweil

The thing that's fascinating about watson, is that it will ONLY get smarter. The more data it takes in, the more examples it sees, the more it understands.

It's interesting watching Watson, especially in the NOVA documentary, where it demonstrates in a game how it can learn. After each question is answered, that information is passed to Watson in a text file. Thus, it learns from its mistakes and knowledge holes, and gradually learns the context of the questions. The only thing it can't do, is that it can't hear incorrect answers, which could help it narrow down it's confidence intervals (the example with Jennings, where it repeated his answer).

Watson is going to be HUGE for IBM, and it will pay off in droves. I can see the END of the phone menu systems that suck ass, and being able to basically ask Watson 3.0 a question, and actually have it respond to what you're saying. Hell, I could see Google getting in on the act, and having the same kind of functionality.

A tool like Watson would make the Google Books project explode into a whole new range of awesome. Given enough input, and enough interaction, in about 20 years, we might see it make a run at the Turing test.

Plot Device said:

My knowledge of AI is limited to movies and a few hypotehtical let's-imagine-the-future magazine articles and TV news stories.

My understanding from this limited knowledge base is that two key developments-- a) self-awareness and b) intuitive conceptualizing -- are necessary before a "mere computer" can make the leap beyond the status of "machine" and into the superior level of sentient independent thought.

And only THEN can our Computer Overlords at last take over.

Therefore, Watson is at the very least embarking upon baby steps toward b) intuitive conceptualizing. But that other all-important criteria of a) self-awareness doesn't seem to be in the cards yet. (Maybe NEXT year, eh?)

Click to expand...

I watched the Nova episode, and I have the impression it's almost a 'trick.'

There are some things it has obviously gotten wrong. Someone mentioned "Which US City" and it answered Toronto. Nova showed a question "This First Lady was born..." and it answered Richard Nixon (who was NEVER a First Lady). The thing apparently didn't have the concepts that "First Lady" is always a woman, and it's hard to imagine how it answered Toronto for a US City.

I found it interesting all the databases it has - various books, local copies of online sites including Wikipedia...

Dommo said:

The thing that's fascinating about watson, is that it will ONLY get smarter. The more data it takes in, the more examples it sees, the more it understands.

It's interesting watching Watson, especially in the NOVA documentary, where it demonstrates in a game how it can learn. After each question is answered, that information is passed to Watson in a text file. Thus, it learns from its mistakes and knowledge holes, and gradually learns the context of the questions. The only thing it can't do, is that it can't hear incorrect answers, which could help it narrow down it's confidence intervals (the example with Jennings, where it repeated his answer).

Watson is going to be HUGE for IBM, and it will pay off in droves. I can see the END of the phone menu systems that suck ass, and being able to basically ask Watson 3.0 a question, and actually have it respond to what you're saying. Hell, I could see Google getting in on the act, and having the same kind of functionality.

A tool like Watson would make the Google Books project explode into a whole new range of awesome. Given enough input, and enough interaction, in about 20 years, we might see it make a run at the Turing test.

Click to expand...

The Turing Test gets back to Plot Device's comment on self awareness. There's no way to determine "true" self awareness. Someday machines might be self-aware, or they just might be doing "incredible simulations" of self awareness. How can you tell the difference?

Shit there's no 100% way to determine if we're self aware, or that we're actually experiencing "objective reality".

I think the truth is, that once an AI can self improve and pass the Turing test, that it won't matter if it's sentient in the sense that WE are. I think that's the primary problem with the thinking of a lot of people. We think of intelligence as having an innate human quality, when the truth is that we really don't know how to define it.

I'm in the camp, that we'll know it when we see it. What I see here in Watson, is potential, and the possibility of being able to untap and comprehend the bulk of human knowledge. Watson definitely doesn't qualify as "intelligent", but given time and examples, I think it can only get smarter.

We humans are currently stuck where we're at. We can teach ourselves, but we're fundamentally governed by the hardware we're born with right now. Watson isn't, and any AI of the future won't be. We may be able to someday enhance our minds, but we'll never have the potential limitless ability for growth that a purely artificial mind could have.

Right now Watson has about a million books worth of information, and has some ability to recognize the context and meaning of the written word. Five years from now, Watson might have the entire corpus of the human written word (and likely a good understanding of the spoken word), and might have accumulated enough experience to directly interpret most human languages without error. I think that's well within the range of the possible.

I don't know if the singularity as a Techno-Rapture will ever come to pass, but I think there is a real possibility that a transhuman future WILL occur. Think ghost in the shell. The determination of what constitutes being alive or sentient will be debatable.

I'm actually okay not being super-intelligent. I'd rather just be me, but hotter. I'll let the super intelligences go and turn Jupiter into a bunch of sentient matter or something.

regdog said:

Why are some humans intent on making us obsolete?

Click to expand...

It's the downward spiral of humanity
without us realizing it.

Dommo said:

Shit there's no 100% way to determine if we're self aware, or that we're actually experiencing "objective reality".

Click to expand...

Sure there is. Step out in front of the next bus you see, while holding the thought in your mind that it's only a mirage. If you survive, we're self aware, and everything exists only in our minds.

Just watched.

Not impressed.

Don't understand the point of a human playing it. The computer is loaded with keywords and then it gets the answer. Big deal.

Cute gadget.

Nothing that much better than Speak and Spell.

Zoombie said:

I'm actually okay not being super-intelligent. I'd rather just be me, but hotter.

Click to expand...

billythrilly7th said:

Just watched.

Not impressed.

Don't understand the point of a human playing it. The commuter is loaded with keywords and then it gets the answer. Big deal.

Cute gadget.

Nothing that much better than Speak and Spell.

Click to expand...

It's probably loaded with a dictionary or two, so yeah, it's got some words in it.

I'm enjoying see good ole Ken again, I'll tell you what. And Watson got something wrong again today. I didn't know about the "confidence levels" and missed the part where he guessed. (Yesterday)