normative
Good AI Answers Require Alignment And Accountability
An AI that gives factually good answers raises two distinct problems: value alignment, meaning the degree it agrees with a society's epistemic norms, and accountability, meaning who gets to correct its mistakes and overcome its biases, which depends on whether the source code, API, and data are participatory or known only to a few.
normativepending
Speaker
Audrey TangEvidence Quote
“The other part to ask is about the accountability, like when it makes mistakes, who gets to correct those mistakes? When it’s biased, who gets to participate in overcoming the bias? Is the source code, is the API, is the data that it uses participatory, or it is known to only a few?”
Created: 6/12/2026, 4:57:43 PM
Your notes
Loading notes...