AANA turns tricky AI behavior into something people can see: generate an answer, check it against constraints, repair it when possible, and measure whether usefulness and responsibility move together.
https://mindbomber.github.io/Alignment-Aware-Neural-Architecture–AANA-/
1 Comment
Current AI tech is so amazing it needs to have a bunch of chains put on it so it doesn’t go completely rabid and destroy everything you give it to play with…
Sounds like a lot of work that could be avoided using traditional non AI search methods, except the industry has wrecked the internet with all the slop.