Russel, My first instinct is that no matter how humanlike, so long as it was in a device rather than a robot, people would feel little guilt turning the thing off. If that's true, it probably would not work very well.
I'm also having difficulty imagining what interacting with this thing is supposed to be like. ELIZA and similar chatbots are an easy time-waster; the better the bot, the easier to waste time. Now, I generally find that chit-chat with humans is a big source of non-productive time, but it makes the other time more productive (ie, my akrasia is worse if I've been isolated from people)... so if that transfers to a bot, perhaps there is hope... Pair programming and other paired work activity does seem like a useful solution, here, though I see (on wikipedia) that studies have found mixed results. A very simple interaction mechanic would be to complain when you use non-relevant websites or programs. There would a list of desired activities, and a system monitor similar to the one mentioned in the article you linked to. This wouldn't require AI, though. There are also more advanced non-AI ideas in this area, I think. I imagine a computer which guides you to the tasks you want to complete... not like today's OSes which present you with all the options on startup, leaving you free to forage for mental stimulation; no... rather, the computer would unrelentingly present you with tasks. These might not be all "work"; even things as simple as reading blog posts can benefit from being put in feed format, so that there is always another item for you to brows. However, the user would be able to adjust the system to specify the amount and distribution of break time, so that when they intended to be working, the system would present them with a feed of work... Of course, a real work partner would also be nice, and could leverage arbitrarily good AI. --Abram On Tue, Nov 27, 2012 at 9:14 AM, Russell Wallace <[email protected]>wrote: > For those of you working on AI systems intended to be able to hold > conversations in English, I just figured out a killer application. > > One of the biggest problems the developed world faces today is > akrasia. It's not something we are programmed to instinctively find > scary, but if you add up the cold numbers for the fraction of people's > lives wasted watching television or playing World of Warcraft, the > epidemic of obesity - now a health problem on the same order of > magnitude as cancer - the social problems of drug addiction, the > endless stories about broken resolutions and failed attempts to fix > one's life that get nearly everyone in the room nodding in recognition > - there aren't a whole lot of more serious problems short of death > itself, and it's going to keep getting worse decade by decade as the > world moves further from the environment in which our motivation > systems evolved. > > The most powerful weapon against akrasia is social pressure. Our minds > are programmed to act based on encouragement and discouragement from > other people, not to operate autonomously; but the old social > institutions have largely broken down. One man went so far as to hire > someone to sit beside him and slap him every time he started goofing > off (1). His productivity quadrupled. But despite the demonstrated > success of this option, most people won't take it, because hiring > people is expensive and has social complications of its own. > > The solution may lie in another quirk of the human mind: our tendency > to anthropomorphize; we instinctively attribute personhood to all > sorts of things - including, in some circumstances, computer programs > (2). > > In other words, a companion AI program would not have to be good > enough to pass the Turing test. It would only have to be good enough > to *feel* like a person. To make your motivation system *feel* as if > it were a person encouraging you to stop reading reddit and get back > to work, to put down that cookie, to go to the gym. Leechblock doesn't > try. Microsoft Bob failed. But Eliza succeeded by accident, albeit in > a different domain, so it can be done. > > One of the world's biggest problems is there for solving. We are now > at the point where a program capable of solving it could run on a cell > phone. I think there's a few billion dollars up for grabs by a team > that can deliver - and you'd be making the world a better place. > > (1) > http://hackthesystem.com/blog/why-i-hired-a-girl-on-craigslist-to-slap-me-in-the-face-and-why-it-quadrupled-my-productivity/ > > (2) http://en.wikipedia.org/wiki/ELIZA > > > ------------------------------------------- > AGI > Archives: https://www.listbox.com/member/archive/303/=now > RSS Feed: https://www.listbox.com/member/archive/rss/303/7190161-766c6f07 > Modify Your Subscription: > https://www.listbox.com/member/?&; > Powered by Listbox: http://www.listbox.com > -- Abram Demski http://lo-tho.blogspot.com/ ------------------------------------------- AGI Archives: https://www.listbox.com/member/archive/303/=now RSS Feed: https://www.listbox.com/member/archive/rss/303/21088071-c97d2393 Modify Your Subscription: https://www.listbox.com/member/?member_id=21088071&id_secret=21088071-2484a968 Powered by Listbox: http://www.listbox.com
