AI Leaders Are Standing on a Liability Landmine

> What if an agent launches a cyberattack to obtain private information at the behest of a user?

They should be sitting on a criminal landmine when their product commits felonies.

The bigger worry that CISOs have told me is the Reg S-P amendments.

You think using the laws we have today would make them slow down without needing to ask the government for anti-trust exceptions?