We can't find the internet
Attempting to reconnect
Something went wrong!
Hang in there while we get back on track
Suresh Venkatasubramanian
Brown CS professor; former OSTP official
ai (1)
ai-ethics (1)
ai-governance (1)
×
ai-policy (1)
ai-regulation (1)
ai-risk (1)
ai-safety (1)
transparency (1)
trust-in-ai (1)
Top
New
-
Should AI systems above a certain capability threshold be required to have interpretable decision-making processes?
Suresh Venkatasubramanian strongly agrees and says:
Congress should demand that these systems be explainable. How that plays out will be a matter for innovation. (2023) source Unverified