AI ethics terminology from MIT Technology Review

  • Topic: AI 
  • Thread starter Tom.G
  • Start date
  • Tags Tags
    Ai Ethics
5 replies · 2K views · last reply
Science Advisor
Gold Member
Messages
5,661
Reaction score
4,509
Here are a few excerpts from an article I stumbled across on
https://www.technologyreview.com/2021/04/13/1022568/big-tech-ai-ethics-guide/

The complete list of definitions runs about 50 items. It is sadly true to some extent, and a bit of a change-of-pace from this site.

audit (n) - A review that you pay someone else to do of your company or AI system so that you appear more transparent without needing to change anything. See impact assessment.

diversity, equity, and inclusion (ph) - The act of hiring engineers and researchers from marginalized groups so you can parade them around to the public. If they challenge the status quo, fire them.

impact assessment (ph) - A review that you do yourself of your company or AI system to show your willingness to consider its downsides without changing anything. See audit.
 
  • Like
Likes   Reactions: Keith_McClary and 256bits
I notice more often now that when something 'screws up' , there is usually a 'contractor' around who wasn't following 'protocol'. The solution has been that the business agreement has been 'terminated', thus making everyone feel more secure and welcome using the service.
 
256bits said:
there is usually a 'contractor'
"Uberization"?

You just need to train a classifier on ethical and unethical ... .
xkcd-style comic: a stick figure presenter pointing at a bar chart on a projector screen, saying the AI methodology was checked by a classifier trained on good and bad methodology sections, which says it is fine.


2451
 
  • Like
  • Haha
Likes   Reactions: 256bits, Tom.G and collinsmark
The bitter biting rapier-like wit of the authors obscures fundamental problems. Can one expect ethical behavior without morality? How do we replace morals based on superstition and tradition with reason?