Data science at American Express
Weaving explain-ability and building trust in AI systems
“Statistics are like bikinis. What they reveal is suggestive, but what they conceal is vital.”
― Aaron Levenstein
How do I know when to trust AI,and when not to?
Who goes to jail if a self driving car kills someone tomorrow?
Do you know scientists say people will believe anything that starts with 'scientists say'.
Designing AI systems is also an exercise in critical thinking because an AI is only as good as its creator.This talk is for discussions like these,and more.
Any scientifically designed system is only as good or as bad as its creator.
At the end of this session,users walk away with an understanding of appreciation of how human thought process influences an AI design process.They shall also be able to critically evaluate an existing AI implementation and weigh on its pros and cons without getting into the inner workings of the actual algorithm.
Finally,the next time someone tries to tell them 9 out of 10 dentists recommend brand X,they know what they're hearing is only a part of the bigger picture.
[Slide links not available yet.I will update when available.
Anyone who has oodles of curiosity is welcome.For the hands-on session,basic familiarity with python and ML would be good to have.