While I was initially impressed with it's context window, I got so sick of fighting with Claude about what it was allowed to answer I quit my subscription after

While I was initially impressed with it's context window, I got so sick of fighting with Claude about what it was allowed to answer I quit my subscription after 3 months. Their whole policing AI models stance is commendable but ultimately renders their tools useless. It actually started arguing with me about whether it was allowed to help implement a github repository's code as it might be copywritten... it was MIT licensed open source from Google :/