Alternatively, could we consider treating the technology as a potentially dangerous technology without giving politicos all over the world additional ammunition

Alternatively, could we consider treating the technology as a potentially dangerous technology without giving politicos all over the world additional ammunition and incorrect mental models with regards to the regulation thereof? To be clear - I fully support treaties banning the use of AI for autonomous warfare. But modeling all applications of AI under the "weapon" label is a massive category error.