http://www.npr.org/sections/money/2015/05/20/406484294/an-npr-reporter-raced-a-machine-to-write-a-news-story-who-won The machine is faster and technically correct, but the reporter still win in terms of making a better piece.
I don't think it's much of a victory when it takes a reporter with decades of experience and a familiarity with the subject to beat the AI, and even then, the AI still churns out the article faster than the human.
>... the reporter still win in terms of making a better piece. I doubt that'll last long.
If you'll pardon answering your question with other questions... Isn't each new version of software a bit better than the last...less buggy? Anyway, that's the plan. How would a journalist algo be any different? That's as specific as I can get.
His point is it's a matter of time regardless, wouldnt you agree?
nope. AI has been "just around the corner" since 20 years before I was born. It's only been in the past 10 years that the professionals have begun saying "that which you consider AI may never happen." I love me some SubredditSimulator but Markov chains do not a conversationalist make.
Searl's Minds, brains and programs, the source of the "Chinese room" argument you've no doubt seen many times, was published in 1980, The Emperor's New Mind in 1989. There have been people dismissing strong AI since there have been enough people playing with computers for it to matter. You didn't hear much from them before because modest uses of AI don't make interesting pop science. What changed is that the singularity nerd religion got big enough that people feel obliged to burst bubbles publicly rather than just arguing with each other.It's only been in the past 10 years that the professionals have begun saying "that which you consider AI may never happen."
I doubt I mentioned it, just because the argument involves quantum physics and Godel's theorems and I avoid conversations on the Internet involving minds and either of those things. Penrose's books are pretty popular in the same circles Martin Gardner and Douglas Hofstadter's books are, though.
...it's all about going exponential. Hang on, Sloopy.
Call me whatever you want, but until I actually see it in action, I'll be triply doubtful - one because it's Chinese (and to me China doesn't have such a good track record of new tech in the industrial era), two because it's Tencent (which I know about - and unless it saved them massive amounts of money, they wouldn't have paid for the R&D) and three because of how simply... a stretch it is. Medium-sized "perfect" news story generated in 60 seconds? IF they did do that, just how much processing power did they put behind it?
It was written in Chinese and completed in just one minute by Dreamwriter, a Tencent-designed robot journalist that apparently has few problems covering basic financial news. Okay. Let's be a bit fair here. I could probably write a financial news article and I can't write prose to save my life. Still, it's an impressive first step. How long will it be until computers start writing the narrative for a nature documentary about nudibranches or a feel good news story about a little girl with cancer and her doll collection?“The piece is very readable. I can’t even tell it wasn’t written by a person,” said Li Wei, a reporter based in the southern Chinese manufacturing boomtown of Shenzhen.
Could you write that financial article in 60 seconds tho?
I expect one day we'll all wake up to an autonomous economy, the whole thing just having sneaked up on everyone. Why don't you post that?
Yes. If the editor is willing to take . . . "Shit's going great over here guys. We're making money hand over fist and we have a merger in the works with those dudes down the way. We will let you know more next quarter so sit tight because we have a few more surprises up our sleeves."