At 12:13 17.11.02 -0500, Ben wrote:
>
>* time to development
>* cost of development [more time and more expertise are needed]
>
>Thus, commercial deployment of AGI or even AGI-ish technologies is naturally
>going to be limited to areas where the quality of behavior provided by
>narrow AI systems just isn't satisfactory.

We did standard software development, but we found problems we can't solve,
but with narrow-AI there are easy solutions. Now we see problems, we can't
solve with narrow-AI, we are using few AGI-ish tools. ...

>In the case of bioinformatics, for instance, I believe there are a lot of
>highly pertinent patterns in biological data that narrow AI methods are not
>seeing.  Having software that can see these patterns is of great commercial
>value.  How full of an AGI one needs for this isn't clear at this point, but
>it's clear that one needs something with vastly more general intelligence
>than current narrow AI software.

We use the same tools for spam-detection and cancel-bots: Clustering, SVMs,
decision trees. 

It reduced our hardware-costs by half.

>I find it quite credible that for your business at the moment, the time and
>financial cost of going in an AGI direction doesn't make business sense.  If
>you had funding to just "build stuff" for 3-5 years and then launch a
>product and worry about selling it, it would be a different story.

We are developing QA-add-ons for our chat-software, we need semi-automatic
knowledge extraction. We have 1-2 years to build qa-stuff, imho its
AGI-ish. We use CLIPS, NARS-ideas and MuliNet. Its slow, today 30s/sentence
on 1 GHz PC.

Thats no standalone product to sell, its an add-on for existing software,
so its much easier to get the invested money back.

>I have doubts about how "robust" you can make narrow-AI chat software.
>Unlike narrow-AI chess software, I believe narrow-AI chat software is always
>going to basically suck.  However, it may serve some specific
>narrowly-defined business needs well, in spite of its suckage judged by the
>standards of flexible human conversation...

Look into community-chats and you'll see how real human conversation sucks.
Look into bestselling newspaper and its not much better. With chatter-bot
we'll get a paid plattform to experiment with narrow-AI and AGI-ish-Software.

We tested a dirty talk bot one year ago, the users wrote us thank-emails. 

Look at Dadaism, very funny or http://slate.msn.com/?id=76886 (The Complete
Bushisms) much more fun.

cu Alex





-------
To unsubscribe, change your address, or temporarily deactivate your subscription, 
please go to http://v2.listbox.com/member/

Reply via email to