AI may write a great restaurant review or local experience story, but the lack of fact-checking is disturbing and I hope the corporations behind these tools can be held to account when they inevitably harm someone with egregious misinformation.
AI may write a great restaurant review or local experience story, but the lack of fact-checking is disturbing and I hope the corporations behind these tools can be held to account when they inevitably harm someone with egregious misinformation.
Isn't the deception itself a sufficiently egregious harm?
Personifying it this way -- finding someone who is clearly injured -- is the only way to make progress.
Thanks for replying. I wonder, although, in US America for example, it requires "demonstrable harm" to bring lawsuits etc, there is an enduring concept of "the fabric of society" [0] in many legal traditions. (I think it is likely examined by J.S. Mill) and even appears in US congressional commentary [1]. In a digital context there is some mileage here [3].
Might we see a successful lawsuit on the merits that the "mere existence" of some technologies is an affront. "Artificial Intelligence" is after all "artifice"
artifice: a clever trick, guile, deception, cunning, a skillful or artful contrivance
[0] https://societyforpeace.com/fabric-of-society-meaning/
[1] https://www.congress.gov/bill/118th-congress/house-bill/1027...
[3] https://digitalcommons.law.seattleu.edu/sulr/vol27/iss1/2/