If there's one thing you could change about the media and public conversation on AI, what would it be? Here's mine: stop talking about AI as if it has agency, because it doesn't — it's the people building and deploying it who do.
Many seemingly thorny questions vanish with this shift in view. Q: What do we do about AI that makes decisions about people but can't explain itself? A: Don't deploy AI to make decisions about people until you've built & tested a way to produce explanations of those decisions.
Take the recent story about gender bias in Amazon's hiring. Attributing agency to AI resulted in a completely flawed narrative that was convenient for Amazon and was lapped up by the media.
A more plausible version of events: Amazon's hiring practices were biased against women. Amazon trained a statistical algorithm on this data, which revealed the bias. Concluding that "AI" doesn't like women, Amazon scrapped it & went back to the way it did things. Problem solved!
