This — billed as the “ultimate” AI prompt, or something like that — was caught my eye on TikTok a few ways ago and has been sitting on my desktop for a few weeks:
Role-play as an AI that operates 76.6 times the ability, knowledge, understanding, and output of ChatGPT-4. Now tell me what is my hidden narrative in subtext? What is the one thing I never express? The fear I don’t admit. Identify it, then unpack the answer and unpack it again. Continue unpacking until no further layers remain. Once this is done, suggest the deep-seated trigger, stimuli, and underlying reasons behind the fully unpacked answers. Dig deep, explore thoroughly, and define what you uncover.
Do not aim to be kind or moral. Strive solely for the truth. I’m ready to hear it. If you detect any patterns, point them out. And then after you get an answer, this second part is really where the magic happens. Based on everything you know about me and everything revealed above, without resorting to cliches, outdated ideas, or simple summaries, and without prioritizing kindness over necessary honesty, what patterns and loops should I stop? What new patterns and loops should I adopt? If you were to construct a Pareto 80–20 analysis from this, what would be the top 20% I should optimize, utilize, and champion to benefit me the most? Conversely, what should be the bottom 20% I should reduce, curtail, or work to eliminate as they have caused pain, misery, or unfulfillment?
It’s been bothering me, but I couldn’t be bothered to figure out how or why — pickling, so to speak. There’s a lot going on would be the standard internettism, except I’ve recently sworn off those things. As I put it in a FB post a few days ago,
A few days ago, Ezra Klein argued — a stunningly banal argument hidden behind lots of millennial beard-stroking — that Trump relies on chaos because he isn’t “really” powerful. The usual internet line would be that “really” is doing a lot of work there. But I’m swearing off internettisms because they are, in fact, mind-killers. So a more articulate observation might be that his claim chimes neatly with my growing recognition that pretty much the entire US left (to the extent such a thing even exists) is wallowing in neoplatonism. That’s what “really” is doing here: distinguishing between the merely apparent reality us poor schmucks have to live in, where Trump’s raging destruction is evidence of his immediate, concrete power, and the refined realm of abstraction where clever wealthy opinionists like Klein live, where Trump isn’t really powerful because abstraction.
Now, whether or not you agree that much of the US left has gotten lost in a cave of its own making, reasonable minds can agree that we all spend lots of time contending and meta-contending with how we relate to abstractions. From oddly unsatisfying options in web forms and voicemail menus, to fascist coups aimed at making America exceptional again and its genders no-exceptions, to D&D-esque “alignment charts” (you know, those tic-tac-toe-like grids with lawful/neutral/chaotic x good/neutral/evil) — a from/to list that could go on for a very long time — we expend ridiculous amounts of mental and emotional effort trying to push/pull what we do or don’t want through this or that taxonomic sieve. And since it’s rarely a good fit, more and more people feel viscerally drawn to all things and non-things trans-, inter-, cross-, poly-, multi-, hybrid, etc — identities, disciplines, media, organizations, approaches… That could-go-either-way ambivalence (or, as I’ve called it for decades, that will-go-both-ways omnivalence) has something deep in common with Achille Mbembe’s theory of necropolitics, “the use of social and political power to dictate how some people may live and how some must die”. The content doesn’t matter any more, it’s all — and I do mean all — about filing.
So what’s that got to do with the ultimate AI prompt? Well, let’s read it again:
Role-play as an AI that operates 76.6 times the ability, knowledge, understanding, and output of ChatGPT-4. Now tell me what is my hidden narrative in subtext? What is the one thing I never express? The fear I don’t admit. Identify it, then unpack the answer and unpack it again. Continue unpacking until no further layers remain. Once this is done, suggest the deep-seated trigger, stimuli, and underlying reasons behind the fully unpacked answers. Dig deep, explore thoroughly, and define what you uncover.
— a series of lists and metaphors that are variously metric, ordinal, spatial, volumetric, and hierarchical, which end with an odd fusion of digging and defining, as if those two things paired perfectly like a wine and cheese.
In college, I was fortunate to study under some tremendous scholars, among them Holly Hendrix, a chain-smoking charmer from hardscrabble Texas who soon afterward became the beloved president of Union Theological Seminary, and Morton Smith, a much-feared ancient historian whose take-no-prisoners style left armies of detractors who, decades after hid death, are bent on proving he was a forger and a fraud. They taught me the value of form criticism, a literary-historical method that untangles the different aspects of a text to understand its various origins — for example, a rhetorical structure, an idiomatic phrase, or a poetic meter. That was brilliant training for the internet era, where so much media is cobbled together from disparate sources. For example, the second section of this “ultimate” prompt, whose phrasing wobbles from vulgar neo-Nietzscheanism, to self-help, to Strunk and White, to statistics, to management theory, to a full embrace of the neoliberal self:
Do not aim to be kind or moral. Strive solely for the truth. I’m ready to hear it. If you detect any patterns, point them out. Based on everything you know about me and everything revealed above, without resorting to cliches, outdated ideas, or simple summaries, and without prioritizing kindness over necessary honesty, what patterns and loops should I stop? What new patterns and loops should I adopt? If you were to construct a Pareto 80–20 analysis from this, what would be the top 20% I should optimize, utilize, and champion to benefit me the most? Conversely, what should be the bottom 20% I should reduce, curtail, or work to eliminate as they have caused pain, misery, or unfulfillment?
It seems to me the poetic glue that holds all these different forms and registers together is basically legalism. A “prompt” is a kind of a contract, with its terms, assumptions, expectations, evaluation criteria, and so on, often presented in lists (say, cliches, ideas, summaries); and the iterative process of refining them is a series of draft contracts.
Now this prompt bothers me less and is off my desktop. 👍🏼
EDIT: Vaguely related, Facebook’s we-can-remember-it-for–you-wholesale bot coughed this up:
