What could a well-known New York comedy mystery possibly have in common with New York AI legislation? Drawing from Hulu hit "OnlyMurdersInTheBuilding", I savagely (ahem) expose the inadequacies of the latest New York #AIsafety legislation, the #RAISEAct.
As the "Only Murders" character Charles Savage put it, "there's a thin line between civilisation and chaos". Have we really absorbed the lessons from #nuclearsafety, or from global warming lock-in? Cinda Canning mused, "sometimes a second chance is just another chance to get it wrong."
The RAISE Act fits a worrying pattern in US AI safety legislature – namely, there is too little to adequately prevent catastrophic harm, while corporate interests are protected.
This post argues for the New York RAISE Act to be amended to include a mandatory independent assessment regime, and suggests how that regime should be structured using the sardonic "Only Murders" detective's crime-solving methodology! #AIgovernance
Read more at the link below.
https://www.lesswrong.com/posts/DniEo2DWXjkt46ihF/solving-the-raise-act-like-a-fictional-new-york-detective
