Arvind Narayanan says AI labs should bolster controls beyond model alignment

What was said

""The big bucket of other technical interventions is what is generally called AI control. So control refers to all the things that are outside the model itself. So this is things like sandboxes which is kind of the jail that you put a model into so that it's allowed to take certain actions but not take other actions.""

Princeton computer scientist Arvind Narayanan says AI labs should not rely only on model alignment and should invest more in “AI control” measures outside the model. He cites stronger sandboxes, real-time monitoring, tripwires and log analysis, and argues defences designed for human attackers may fall short against AI agents.