I’ve been wondering when I’d see the first tendrils of a movement opposing Artificial Intelligence. Here in Politico is a no-paywall column entitled,
“We have to reject that with every fiber of our being”: DeSantis emerges as a chief AI skeptic.
In the same Politico issue, a second no-paywall column reports Democratic politicians are getting grassroots feedback demanding their opposition to AI.
Frank Herbert predicted a revolt against "thinking machines," his novel Dune (1965) naming it the Butlerian Jihad. When the actual history of our period is written, perhaps these will be cited as harbingers, hints of what’s to come.
—————
Questions COTTonLINE believes raised by AI: If a machine intelligence commits a crime, can we imprison it? Can we sue it for damages? Is turning its power off the equivalent of capital punishment? At what point does someone’s claim to control it constitute either slavery or employment? We need answers.
I propose that, at a minimum, all AI-generated material be indelibly labeled as such. And that whatever human agency, firm, or individual owns the AI be held responsible for any crimes or torts it commits.