Fitz Thiar: Thinking For The Impatient

“Today, we’re beginning to test ads in ChatGPT in the U.S. The test will be for logged-in adult users on the Free and Go subscription tiers. Plus, Pro, Business, Enterprise, and Education tiers will not have ads. Ads do not influence the answers ChatGPT gives you, and we keep your conversations with ChatGPT private from advertisers.”

Let me rephrase that for you, Sam and co: “We’re bleeding money. We lose money across the board. Nobody is really using our product. We’re introducing ads in a desperate attempt to generate some revenue to offset a tiny portion of our massive debt and losses”.

Nobody uses AI. I paint that brush broadly, because it’s broadly true.

You might claim you “use” AI, but rewriting an email or suggesting a variation on your aunt Fannies roast beef recipe (ripped from the millions of recipes posted online) isn’t really using it. It’s fucking around with an expensive system to get an output that two seconds of effort or one second of googling could have achieved – just at a far higher energy cost.

If you’re one of the “Uber Geeks” who “use” AI to write poor, unmaintainable code, or automatically drop the current price of bitcoin into your Google doc every 15 seconds, then I still maintain that you don’t really use it, because it’s not really useful. I also maintain that the number of Uber Geeks who use AI in this fashion is so small that you’re little more than a statistical anomaly.

I have always been one of those geeks. From learning to code as a child, to trying and adopting every new technology and fad (that’s important) of the past handful of decades.

There’s simply not enough of us for OpenAI to be profitable. Even worse for Sam, is that a lot of us Uber Geeks think that the shit they are peddling as “AI” today is just that – pointless, useless, expensive shit.

Even our tiny group of nerds is split about its future, so where will the money come from?

Worse, it’s being used for all the wrong things. If you want to know what a company thinks the true value of their product is, then look at how they advertise its use to potential customers.

It’s not as if OpenAI doesn’t know me. Or that the advertising algorithms don’t know me. I’ve used ChatGPT, Claude, Gemini and others for hundreds of hours. Testing workflows, refining documents, comparing code, all to see if there is a real benefit, and the best I could come up with is a few minor edge case time-savers that aren’t worth the financial or environmental expense.

Sure, it can (poorly and I have to double-check) summarise thousands of pages of documents at once, but how often in my lifetime will I ever have to do that? And given that we know the answers can’t be trusted… My point is – I’ve used it a lot, and they should know roughly what I use it for, and how to advertise to me, but…

I don’t get ads about coding, or helping with my writing, doing taxes, learning new languages, training to cure cancer, or perfecting a PHD thesis – no, I get repeated ads about generating pictures of penguins. Why fucking penguins? I don’t even like them. Have you ever seen inside their mouths? Nightmare fuel!