Resisting AI

For a while I would not go anywhere without my headphones throughout the summer holidays. Probably I was already too overwhelmed with my kids at home. I still listen to the radio in the morning after dropping my wee one at school and walking the dog. But I guess I listen to a lot less podcast.

I listened to Cal Newport’s conversation with Brad Stulberg about resisting AI. I haven’t read Cal or Brad’s books but I think they are both American? They write rather self-help books? I don’t mind reading/ listening to self-help once in a while, although I am highly skeptical of people who only read self-help books. I am not sure if I want to live an excellent life, I probably am quite happy with a mediocre life – but what else is there for me?

As I keep talking myself down and feel like I am at a crossroad everyday, I do ask AI what should I do with this and what should I do with that, I even ask AI how can I be more confident as my upcoming essay forbids the use of AI. I ask AI questions and whatever I ask it is very encouraging and supportive, telling me that I am on the right track or nearly, but I don’t believe LLM is capable of genuine assessment of whether I am doing fine or not. I have to admit that it is not a very healthy relationship. And I really don’t like how it has to end every conversation with ‘if you want, I can suggest 3 methods of/ a refined version of’ etc. But there are so much about people I don’t like as well, so it is not too bad either. I don’t got people to talk to, you see, not at work, not in my study, not with my family. AI dialogue is artificial by name, artificial by nature, but is it better than nothing? I am not sure.

My experience of asking AI to help me to writing reminds me of two things – 1. hanging on the edge of the swimming pool and too scared to let go; 2. putting a piece of paper over comic books to copy and think that’s my drawing. I suppose Cal and Brad are right that if I want to be able to write – or make effective decision, or understand why rationalist approach to policy-making rarely works, or be a effective student, I need to figure it out myself. I am going to write to myself on paper and answer myself instead.

Anything worth doing, you have to figure out yourself, however clumsily and slowly.