(Photo by Senwart Machinery on Unsplash)

You've seen the recent fallout over genAI cost overruns. You may think the cure to tokenmaxxing would be tokenminning.* Not quite. The cure is project pruning.

Try to see your token budget as a financial budget that you can allocate to different projects.

Some projects certainly need those tokens to reach their true potential. But others are dead weight that aren't going anywhere.

Many of you are wasting LLM tokens on that latter category!

Take a deep, honest look at your genAI usage and you'll see what I mean.

You'd certainly kick a loser stock pick out of your investments portfolio; why are you holding on to dead-end genAI projects? That makes no sense. And you're wasting time, effort, and money.

If you need to cut your AI spend, start by pruning the dead-end projects.

(*== Yes, I just made up the term "tokenminning." )