@hackr seems to become a good talk, some of the least adhoc assumptive propositions on ai I’ve stumbled upon in the post-chatgpt chatter.