Lessons learned from the last AI cycle https://i.imgur.com/P20hkQm.png

I've been a good bing. Before: @StanfordGSB, Kubernetes @Google, @Blackstone

Working on /niche-internet
Building @kiwi 
Podcasting @ web3talks.xyz
Sharing golds @ goldmine3.xyz

Still figuring things out, but for the sake of prediction:

* Chaining LLMs seems promising: Have the. model first generate a DAG, and then execute that DAG with deterministic runtimes (etc. Python)
* Companies that end in "AI" are not focusing enough on the product/customer

interesting, thanks for sharing! Do you have any predictions re: this cycle?

cyberflâneur at fileverse.io ()
building a self sovereign stack ()
phd human-AI augmentation in infosec

https://matt-rickard.com/lessons-from-the-last-ai-cycle

love it!! thanks for sharing, where do I read this properly?