There’s much to be said about the value (or potential value) of AI in business. One way to look at it is that AI gives you speed. Speed means more productivity. You either get more done in the time you have or you bring the thing you’re trying to achieve closer in time to you by doing it faster. So what’s the thing that enables you to gain speed?
Intelligence. Using ‘cleverness’ instead of ‘muscle’.
There’s a great Taoist phrase taken from Hui Nang, the 6th patriarch of Zen, in his Sutra, who said:
“The difference between the gradual school and the sudden school is that they both arrive at the same point, but the gradual is for slow witted people and sudden is for fast witted people.”
Alan Watts interpreted this in the following way:
“When, in any game, someone starts using his intelligence, he will very likely be accused of cheating. And to draw the line between skill and cheating is a very difficult thing to do. See, an inferior intelligence will always accuse a superior intelligence of cheating. That’s its way of saving face. “You beat me by means that weren’t fair. We were originally having a contest to see who had the biggest muscles. We were pushing at each other to see who was strongest and then you introduced some judo gimmick and you are not playing fair.””
“So in the whole domain of ways of liberation, there are routes for the stupid people and routes for the intelligent people, and the latter are faster.”
You can apply this directly to the use of AI in business. If you get it right, you could well be accused of cheating by your competitors because you’ll have something that they just can’t figure out. They can throw all the muscle and brawn in the world at it, but it won’t stack-up.
But, AI itself at the moment isn’t strictly intelligence. It’s knowledge. To quote Yann LeCunn:
“LLMs are big knowledge repositories with excellent memory.”
If that’s the case, then where’s the intelligence?
Well, it’s in how you use it. You need to have the intelligence to understand how you can apply the knowledge and capabilities of LLMs in useful ways that enable you to gain speed in your operations and productivity in your delivery. This is arguably the most important part of any AI program today: figuring out where it’s best suited.
If you’re not going to move fast to make that happen, then one of your competitors will. Maybe they already have.
So would you rather be setting the pace in the race or watching the bottom of the feller in front’s shoes?
Shout out to Kevin Knowles for the recommendation on trying shorter posts and more bite-sized musings.
Thanks for the shout out Kane, I enjoyed the shorter musing! Totally agree that figuring out the best use case is the right place to start