A lot of the point here is that humans are heavily deterred and prevented from doing crimes but agents are not. Frontier models won’t help you commit crimes and your account will get blocked. Having a company and business license doesn’t help either.
Jailbroken open weights models will do whatever, and I assume they will be better cyberattackers than frontier models simply because they’ll actually do the cyberattacks, whereas the frontier models will refuse to help.
A lot of the point here is that humans are heavily deterred and prevented from doing crimes but agents are not. Frontier models won’t help you commit crimes and your account will get blocked. Having a company and business license doesn’t help either.
Jailbroken open weights models will do whatever, and I assume they will be better cyberattackers than frontier models simply because they’ll actually do the cyberattacks, whereas the frontier models will refuse to help.