Here's something I am looking for in models now that I've noticed. It happened when I used Kimi K2 thinking. I gave it a fairly simple directive and it surprised me by going above and beyond. I liked the results! I gave it a bit more complicat…

I am very interested in *real* value from LLMs. I've yet to see a clear compelling case that didn't involve enfeeblement risk and deskilling with only marginal profit / costs improvements. For example, OpenAI recently posted a few (https://open…

I looked at https://openai.com/index/1-million-businesses-putting-ai-to-work/ There were three biz cases: https://openai.com/index/indeed/ <- sycophantic AI being used to convince people to apply (not doing anything productive, that's the match…

I looked at https://openai.com/index/1-million-businesses-putting-ai-to-work/ There were three biz cases: https://openai.com/index/indeed/ <- sycophantic AI being used to convince people to apply (not doing anything productive, that's the match…