3 ms·
You've moved the goalposts from, "we should assume that predictive text models are sentient," to, "we should be wary of programs explicitly designed to interact
by cjohnson318 4y ago
You've moved the goalposts from, "we should assume that predictive text models are sentient," to, "we should be wary of programs explicitly designed to interact unpredictably with other computer systems," something that everyone from the Amish to the NSA to Cozy Bear would agree with.
- doctor_eval 4y agoWhat I said at the outset was, > Considering that we don’t know how the brain works so well, and we don’t understand why LLMs work so well, simply on the basis of their output I think the safest assumption is that these models do indeed have agency, or at least the capability of agency. These systems can already interact with others, it’s not moving the goalposts, it’s common knowledge. Anyone with access to the APIs can make it happen. Or are you now claiming we are just talking about one specific LLM and not LLMs generally? Anyway this discussion is fruitless, I’m out.