Becky Smith

The proportion of people representing themselves in court has risen in the advent of AI. Prior to the launch of ChatGPT in 2022, the proportion of pro se (“for oneself”) civil court cases was stable at around 11 percent. Since then, it has risen to nearly 20 percent. Moreover, the number of complaints that have been seemingly AI-generated has also increased, from “essentially zero” to, again, nearly 20 percent this year. The watchword of the past decade of tech development has been individuality. The promise was that technology would free you from the mediating structures of collective life - the institutions, the gatekeepers, the slow grinding machinery of things that existed before you did - and return you to yourself, pure and optimised. Artificial intelligence is the steroid of this promise. What better demonstration than a new age of ‘self-representation’? Those at the higher-end of the tax bracket are taking it to the nth degree, using AI to represent themselves - literally. Semafor recently profiled Bill Nguyen, a guy who has “made a small fortune selling multiple companies to Apple” and uses a personal AI which has “all but taken over his life.” The difference is that, unlike the vast numbers of people using the free version of these tools (and, as a result, absolutely shedding money for these companies because it’s definitely not a bubble, folks), his custom AI is actually supposed to be him. Nguyen is “plunging tens of thousands of dollars” into a machine that has access to his Slack, calendars, call logs, location history, and his voice. It now autonomously sets out his day, books meetings for him that he doesn’t know about, and is ‘helping’ raise his kids. A recent discussion between Nguyen and one of his sons about the war in Iran spurred the machine to plan a trip to a museum, unprompted, that tied together current events and ancient Greek and Roman history, an area of interest for his son. I don’t know why you would need a thousands-of-dollars per month machine to tell you that you should go to a museum with your son about a topic he seems interested in, but I digress. Actually, I don’t digress. Mark Zuckerberg is reportedly training his own AI clone, to automate the tech billionaire’s interaction with his employees. The character is apparently being trained on his mannerisms, tone and publicly available statements, and company strategies. But in these attempts to liberate themselves from the legal system, their employees, or their relationships, respectively, these men are losing their authenticity to the machine. The existentialist Martin Heidegger, in Being and Time, described authenticity as Eigentlichkeit, which translates to a kind of ‘ownedness’. It is a perpetual effort, not falling through life through the path of least resistance - and is antithetical to the promises of this ilk of tech products, which offer only the path of least resistance. You can lose the possession of yourself, Heidegger says, then to a kind of anonymous “They”, where “everyone is the other, and no one is himself.” And what greater manifestation of ‘everyone’ is there, than the machine trained on everything anyone has ever put on the internet? These systems are black boxes, their workings unknown, and the motivations of the men who run them equally abyssal. In a recent profile of Sam Altman in the New Yorker, the head of OpenAI was described as someone who “just tells people what they want to hear.” I can’t predict what will happen as people continually take advice from this endless averaging, this hall of mirrors that reflects the person back at themselves, again and again. But these purported gains of agency are, to me, diminishing it - and as they continue down the path of least resistance, the destination is someone else’s idea of who they are. These people need to own themselves - before they get owned by somebody else.