Last week I spoke with two different CFOs, and I can’t stop thinking about these two quotes:
“Whenever anyone talks to me about AI, they talk in such generic terms. It’s not helpful. I was excited to attend this 45-minute AI presentation at a conference, but it was written by AI. It was all the basics. It was nothing specific. I want to know specifically, as a CFO, what I can use, how would it work, what will it save me.”
“Anyone who tells you they have a complete understanding of AI and what the best prompts are for Claude and ChatGPT is lying because it's changing so fast. I can’t keep up.”
Starting simple and generic is useless. Attempting to master an ever-changing concept is impossible. Both of these guys are sharp, so if the CFO isn’t going to figure out AI, who can?
They’re describing two sides of the same problem. There’s a massive gap between legacy businesses and frontier labs, and it might still be getting wider. It feels like we’re all stuck on Earth watching five companies run circles around us on the moon. It’s there, we just can’t get it.
I guess I’m the fool for thinking that I can somehow write about AI without being overly basic nor too complex. If my two readers are these CFOs, my goal is to try not to piss off either of them.
Here’s my promise to both of them and to anyone else reading. Actually, there’s three promises:
Even though it may feel like we’re discussing otherworldly topics, I’ll root each blog in the day-to-day conversations I’m having with businesses.
Every blog has something quick and actionable for someone to apply to their business.
I’ll never write with AI. All sentences are mine, so excuse the typos and lack of em dashes
I’ve honored #1 and #3 already, so here’s the action for those who made it this far:
Every time you say or hear the word “AI” tomorrow, write down the phrase that could replace it. At the end of the day, look at the list.
Why? People are hiding behind the term “AI’, instead of saying what they actually mean. This directly contributes to the problem listed above. The term has become a catch-all phrase that is both too generic and too hard to understand. Sound familiar?
Shifting from “let’s AI that” to “let’s use sales data to determine which prospects to target” is a nice example. Start defining how you want to use the technology before trying to understand it.
Plug your list into Claude and ask “here’s how I’ve been using AI. What am I missing or not thinking about?”
I live two lives. I’m an MBA student at Stanford. I’m also the co-founder of Mavnox, an AI implementation startup for mid-sized services businesses.
Subscribe to get the best AI ideas I come across each week.
-TW