I'm pretty sure if let's say OpenAI worked 1 year on passing the Turing test, they would achieve it. It seems extremely close. Current LLMs are trained to be us

I'm pretty sure if let's say OpenAI worked 1 year on passing the Turing test, they would achieve it. It seems extremely close. Current LLMs are trained to be useful, not to trick humans.