Indirect Normativity Speed Run - Adam Ford & James Hughes #AI #Ethics #future

How can we use AI to help choose good values & norms to guide our future? In Chapter 13 of Superintelligence, Bostrom explicitly distinguishes indirect normativity by its reliance on the AI’s superior processing power to help solve the problem of value. While the other two options like direct specification address the general problem of value selection, they could just as easily refer to "Direct Normativity" (where humans agree to sit down and write the rules themselves). Indirect Normativity moves from humans defining the "good" directly to humans defining a process for an AI to help figure out the "good" for us. Indirect Normativity has a delegation aspect: It highlights that the AI is an active participant in the value "choosing" process. Indirect Normativity takes uncertainty seriously, and puts epistemic humility front and centre: It reflects the idea that we are currently too cognitively limited to get the values right - especially on our first try (before we launch superintelligence). As such Indirect Normativity is a formal procedure use to help guide AI rather than baking in a finished moral code. Many thanks for tuning in! Please support SciFuture by subscribing and sharing! Buy me a coffee? https://buymeacoffee.com/tech101z Have any ideas about people to interview? Want to be notified about future events? Any comments about the STF series? Please fill out this form: https://docs.google.com/forms/d/1mr9PIfq2ZYlQsXRIn5BcLH2onbiSI7g79mOH_AFCdIk/ Kind regards, Adam Ford - Science, Technology & the Future - #SciFuture - http://scifuture.org