Rohan, Eugene, and Shayne talk about what are the technical innovations driving the explosion in Large Language Models recently.
Shayne talks about how much uncertainty there is about which applications LLMs will be good at - uncertainty that might take a decade to be resolved.
Eugene brings his perspective on what a successful seller will look like in a world where synthesis, Mail outreach, summarization, and information retrieval are taken care of by LLMs.
As they continue the conversation - they talk about trusting AI models, and whether we’ll ever move from AI-assisted human selling to human-supervised AI selling.
Rohan, for one, believes that AI will never be able to completely replace the human seller because human beings want to talk to another human when they are making $10K purchases or larger.
Lots of interesting stuff for folks interested in a plain English introduction to LLMs and future applications they might enable.