I am a user of ChatGPT, which I've written about in previous Scene4 essays, and I use it as a sounding board for ideas, a stress-test of my own writing, and an excellent way to take multiple sources and types of information, summarize them, and extract the relevant ideas and phrasing I need for my work.

The AI push these days (really, it should be the AI push this second, so fast do changes and additions happen in the AI galaxy) is toward something called "agentic AI." Put simply, this involves linking AI programs trained to do different things to create "agents" that then assist you in your work. These agents are configured to handle tasks that can range from the repetitive or tedious – scheduling appointments, reading and summarizing an email inbox, booking a trip – to higher-level functions, like screening applicant résumés, investigating cybersecurity threats, or supporting health diagnoses.

The common notion about them is that the agents take the initiative rather than wait to be instructed. The agents show agency, or, as Copilot would describe them, "intelligent systems that understand goals, adapt to feedback, and act with purpose."

Of course, the tech bros who create these things (Altman at OpenAI, Dario Amodei of Anthropic) see nothing but upside in their development, though Amodei did have some dire predictions about how this AI development will make many, many workers superfluous. (He was challenged in this prediction by many people, like Bill Gates and Mark Cuban, who said, yes, the technological revolution will eliminate some jobs but it will more than compensate for the losses by creating new jobs and new types of jobs that will, in the end, be a net plus for civilization – that is to say, the buggy whip makers will go on to work in the new automobile factories.)

These commentators emphasize that handing over the more mundane aspects of work life (scheduling, email readings) to agents frees you up for more creative and productive work, though it is never quite clear what that means and how that would play out in actual work situations. Really, having Sam Altman mouth these words is just the overlord trying to dampen people's legitimate anxiety about the way the world of work (except for perhaps in the upper echelons) is trending toward precarity and contingency, that everyone is becoming, to one degree or another, a gig worker who needs a side hustle to make it - that is, if they can find a job at all. (Pity the 2025 college graduates who have to find work.)

Which is to say what is not really ever said when talking about AI in its various forms: it is a capitalist technology created to serve capitalist ends. And we all know what those ends are because they are what they have always been, what Marx named them almost 200 years ago: it is labor that turns capital into capital and thus into a profit that is extracted by a process of exploitation, and if there can be ways to do that with fewer workers – by, say, agentic AIs – then that is how it will be done. And you know that creativity you'll have more time to exercise once the agents free you up – don't think you actually own that for your own purposes. That goes back into the capitalist matrix, thank you very much.

As Marx pointed out, this exploitation, at least in capitalist societies like ours, is never fiercely slave-like. There will be enough "benefits" hived off to soften the blows so that people continue to continue, fascinated by the glitter just enough to make playing out the old scripts tolerable.

So, while AI is in many ways remarkable as a technology, it is very old hat as an ideology, one that we are going to have foisted upon us without much or any public input about both its gifts and dangers. (Note that the horrific One Big Beautiful Bill has a provision that prohibits for a decade any municipality anywhere in the country from passing laws that regulate AI. The Republicans know all the stops on the gravy train, and the AI money is too big to let anything like public comments derail it.)

What I think is true is that AI won't so much replace people as people who are educated in AI will replace those people who aren't, which forces us to undergo another round of remaking ourselves into whatever it is the capitalist system feels it needs. Exhausting and infuriating, especially when it also means fighting to get one of the increasingly shrinking number of seats in a very vicious game of musical chairs (just shy of Squid Game).

The only bright result I can think of that might come out of this churn is a revived discussion of a universal basic income or some such way of unlinking income from work, especially in the new agentic Eden where so few can create so much wealth while having their calendars updated and plane tickets reserved. I would gladly become superfluous if there was money to be made in it.

But then I give myself a dope slap and realize what society I live in – ease, prosperity, and comfort for all will never trump cruelty, distress, and immiseration for many. Cruelty is often the purpose of the policy, which we can clearly see rising up from the mists of the D.C. swamp.

Or maybe I should set up an agent to run the musical chairs on my behalf while I use my newly released creativity time to foment rebellion and resistance – which I think I will then offload to another agent, of course, because running revolutions can take a lot of time and stress, which will free up even more creative time that I will use to barricade myself against the time when the robots take over and the world ends in fire and ice.

Let me ask ChatGPT to formulate the plan.