Access and Feeds

Aggregated AI: Combining AI Engines into SuperBots

By Dick Weisinger

There have been whirlwind advances by Artificial Intelligence over the past few years. AI that can beat humans at Chess, Go or Jeopardy. AI that can drive cars and drones. AI that is better than radiologists at reading X-Rays and ultrasound scans.

But most AI solutions have been very focused and specific to one problem or limited area. What happens when these point solutions are aggregated into master AI systems that excel at a wide range of capabilities? Or at least the capability for AI engines to intercommunicate and exchange information.

David Linthicum, Chief Cloud Strategy Officer at Deloitte, said that “AI engines are able to combine with your taught models, creating an AI superbrain of sorts, not only providing global experiences outside of your domain but combining with your own training data to provide local and global experiences.”

Taking it even one step further, next- or next-next-generation AI solutions will be capable of more sophisticated AI that can write AI software. Google, Facebook, Microsoft and other tech companies are all researching AI that can build AI tools and models.

Pieter Abbeel, professor at UC Berkeley, said that “Computers are going to invent the algorithms for us, essentially. Algorithms invented by computers can solve many, many problems very quickly — at least that is the hope.”

Digg This
Reddit This
Stumble Now!
Buzz This
Vote on DZone
Share on Facebook
Bookmark this on Delicious
Kick It on DotNetKicks.com
Shout it
Share on LinkedIn
Bookmark this on Technorati
Post on Twitter
Google Buzz (aka. Google Reader)

Leave a Reply

Your email address will not be published. Required fields are marked *

*

three + 1 =