The ancients, who looked to various gods and spirits to guide and punish them or, exact vengeance on their enemies and enforce balance enjoyed a higher quality mythological catalogue than the current set.

Zeus, Huītzilōpōchtli and the notables of countless pantheons, whatever their storied faults, were better entities to think of as being, in at least some sense, real, than modern, deified concepts such as the sober, data driven business person, a figure as mythological as anything depicted on canvas, via marble or on the wall of a cave by dim torchlight but far less nourishing.

A word comes to mind: savvy, followed by a phrase, situational awareness. These are lacking in the modern capitalist who, even when judged by the narrow, profit fixated limits of business, shows little sign of possessing either.

Particularly when it comes to the carnival that is so called ‘AI’.

The promise, from Microsoft, OpenAI and their co-conspirators was that, by surrendering business processes to ‘AI’ workers could be eliminated, thus boosting profits. The tech press, a cheerleading squad about which it’s hard to say a kind word (there are exceptions but their rare appearance only illustrates the problem’s scope) broadcast this yarn, shouting about labour in sectors from fast food to law being replaced by chatbots.

In Greek mythology, the goddess Nemesis ensured retribution for offences against the gods; an enforcement of consequences. Nature cannot be fooled; consequences always arrive and often, precisely as people who pay attention, who are savvy and possess situational awareness, predicted. It doesn’t take special insight to know that stepping into traffic isn’t a good idea, you just have to know how the world works (the not at all theoretical physics of metal, travelling at high speed, impacting your soft body). It should not have come as a surprise to business types that the ‘AI’ they looked to as a means for enhancing their bottom line was, in fact, built by the ‘AI’ vendors for exactly the same purpose.

At some point, well, from the start, really, it should have been obvious, that introductory prices for access to ChatGPT, Claude, GitHub Copilot etc would be replaced by real pricing. Microsoft has bills to pay (data centers aren’t cheap) and profit targets too. It should have been obvious that once you were dependent, the manageable price would disappear, ‘like tears in rain’, replaced by bills that inspired weeping.

A dispatch from Nemesis in the form of a recent CNBC article: ‘Tokens or humans? The new corporate trade-off’

From the article…

The root of the squeeze is that the technology works but doesn’t yet pay for itself.

“The way AI works today, it’s very powerful, but it’s very inefficient,” Jain said. “The value that AI drives at this point is trailing the cost that businesses are incurring.”

A big part of the problem is inefficiency in picking models. Roughly 95% of enterprise AI usage is still running on the most expensive frontier models, even for tasks that could be handled by cheaper alternatives, Jain said.

Thus spoke Arvind Jain, CEO of Glean, an ‘AI’ company. Jain is trying to sell a product whose utility is now being questioned so of course, he presents the problem as being that ‘AI’ is very, very good but too expensive or, you’re doing it wrong. But even for Jain, it’s difficult to edulcorate (a marvellous word, which I learned about recently, that means to sweeten) a souring situation. To say ‘it’s powerful but inefficient’ is akin to declaring a car to be very fast but unable to get you to the store.

In other words: useless.

I started this brief essay by talking about mythologies: sublime and stupid, ancient and modern. In the case of the 21^st century business class, who depict themselves as grim faced gods, staring deeply into luminous screens, they are more clown than Vulcan, prat-falling rather than thinking, chasing after a world without workers and buying magic beans in the hope that world will arrive.

Nemesis is calling.