Current San Francisco discourse:

"He stopped doing his homework! Alvy, tell Dr. Flicker what you read."
"It's about GPT-4."
"GPT-4?"
"If it keeps getting better, we'll have AGI, and that will be the end of everything."

This line in the NYT metaverse piece cracks me up:

As an experiment with prompt engineering, I made a "OaklandBot" chatbot with the leaked FB Llama weights to tell me more about Oakland and it sent me a fake Google doc URL when I asked it a question. Kind of sketchy for an AI chatbot, but pretty on-brand for an Oakland AI chatbot.