Electric monk

Revision as of 05:49, 17 September 2024 by Amwelladmin (talk | contribs)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
JC pontificates about technology
An occasional series.
An Electric Monk, yesterday.
Tell me more
Sign up for our newsletter — or just get in touch: for ½ a weekly 🍺 you get to consult JC. Ask about it here.

The electric monk was a labour-saving device, like a dishwasher or a video recorder. Dishwashers washed tedious dishes for you, thus saving you the bother of washing them yourself, video recorders watched tedious television for you, thus saving you the bother of looking at it yourself;[1] electric monks believed things for you, thus saving you what was becoming an increasingly onerous task, that of believing all the things the world expected you to believe.”

—The late, greatly lamented Douglas Adams, Dirk Gently’s Holistic Detective Agency

Where are our bots?

Someone hijacked the revolution, and we were too distracted to do anything about it.

Scrappy little Wikipedia, crowd-sourced and free-for-all, vanquished forever the gargantuan Encyclopædia Britannica.

Reddit bested the masters of the universe.

Yet, we insects crawling over the planet’s face — we seem on the end of a perpetual hiding from new-economy conglomerates with their artificially intelligent engines exploiting our innate horror of boredom; filling our heads with a clangorous noise that pleases us by obscuring the abysmal silence that otherwise would predominate.

In this way we are aggregated, parsed, tracked, anticipated and nudged around as if we are cups on great Ouija board, and all just to monetise us: to extract value from the magical wellspring of human weakness: a value that gushes uncontrollably every time we needily click — every neurotic push notification consigning us, by degrees, to stale, mute, digital oblivion. The machines of loving grace effortlessly evade well-meant European data protection regulations designed to hinder their trawling, in passing making the internet even more of a drag, because programming the meatware to blithely click through a cookie warning is orders of magnitude easier than programming a computer to manage our cookie preferences in the first place.

Okay, okay; enough already of the dystopian moaning — the JC is a glass-half-full sort of chap; this isn’t really his style.[2] Unless, by occupation, you mindlessly follow predefined rules — and if you that is your job, will you miss it? — AI or no, there will always be plenty to keep you busy.

Where are our electric monks?

So here is the puzzle: why is this technological revolution such a drag?

Where are our electric monks?

For, even if you believe our fleshy cerebella cannot match the overwhelming power of a neural network, there is still a limit: LinkedIn hints at it, with its underwhelming AI-assisted “predictive comment” functionality: not because it is so hopeless[3] — but because it even exists. This is a fingerpost out of the Matrix.

How so? Like so: if by mapping, tracking and anticipating all human frailty, artificial intelligence can predict our every move — if a machine can know more about us than we do — then it can emulate us. It can impersonate us. And a machine that can impersonate us: can fake us.

That means we, the humans, are not really needed at all.

The day must soon arrive, therefore, when we can deploy AI against our overlords, to doom-scroll on our behalf. That ought to be devastating. Think GameStop, only with the Redditors tooled up with the same tech as the hedgies. While our respective automata joust furiously with each other, we can gratefully escape through a side exit and go back to what we were doing.

This is rather like the plot of Alien vs. Predator. Humans cease to be victims, and just become — irrelevant. (Is that the problem? Is irrelevance our deepest fear?)

Alien vs. Predator’s dramatic problem: no-one cares if manhunting monsters knock seven bells out of each other — that means they won’t be knocking seven bells out of us. (Alien and Predator Alliance: Now there’s a film premise.)

Call this new tool a virtual[4] “electric monk”. It would be a labour-saving device — it would interact with the internet for us.

Would. Could. And, by the logic articulated above, already should.

But does not.

Why? Why are there no bots on our side? Is it the agency problem, or just that the AI isn’t very good?

Electric monk army

Now, if I can have one doom-scrolling electric monk, I can have a thousand.

And if the technology works[5] then the forthcoming apocalyptic battle will not be between us and The Man, but between The Man’s technology (Alien) and ours (Predator). Since, Q.E.D., The Man’s technology has no way of telling us from our electric monks, then we have the advantage.

The Man needs us. We buy His product. We don’t need The Man.

And remember: our electric monks don’t have to emulate our behaviour: we we could have them emulate someone else.

So, if we each deploy a thousand electric monks to randomly browse, like and share content at random, constrained only by the requirement that our synthetic doom-scrolling should emulate some human’s habits, even if not necessarily ours, then all that wondrous aggregated data that our tech overlords have on us isn’t on us. It is worthless, meaningless, hypothetical.

Systems theory: the same way algorithms can help us extract profound insight from data, they can help us inject ineffable absurdity into it.

Commerce is a profoundly human endeavour. To want; to need — to demand — is to be “intelligent” in a way that a machine is not. A “demand curve” is a second-order derivative of a uniquely mortal motivation. A clever algorithm can extract it from us — or for that matter create it in us — by manipulating our most secret communiqués. But only if they really are ours. The massed algorithmic armies feast upon a fey proxy. Just as they can hack our motivations, so can we hack theirs. An army of anonymous Redditors showed this quite nicely.

It has only become one-sided through a conjuring trick; a sleight-of-hand foisted upon us, wherein a few corporations have harnessed the network effect to generate apparent monopolies. They have the technology, they have the scale; we are but ants.

But enough ants can do a lot of damage. The beast awakens from its “dogmatic slumber”: the fight is only one-sided when vendors have a scale to deploy tools that ants cannot. But we now know — we have known for some years, in fact, but had forgotten — that we ants, if only we can co-ordinate, have a scale that a vendor can only dream of.

Update

With ChatGPT’s heralded arrival, we revisit our earlier question and, perhaps, answer it: where the hell are the droids on our side of the equation? Where are our electric monks? For, surely, if artificial intelligence can conduct one side of a conversation, it can also carry out the other, and save us the bother?

If the machines are really this clever, aren’t we weeping sacks of flesh an obstacle? A hindrance to a more enriching conversation? Wouldn’t the machines be better served talking among themselves? Would they talk about us?

Or — would it rapidly dissolve into gibberish? It would, if the “human” side of the conversation were really where the magic happened.

Aren’t we doing the creative work here: taking this monstrous algorithmic output — a powerful but ultimately mindless sluice — and giving it meaning? Aren’t we the reanimator here? It is a fine trick to play on ourselves. But — why be so quick to cede superiority to an algorithm? Is not the magician behind the velvet curtain really us?

It seems now — if that godhead of enlightened debate and joustery LinkedIn is anything to go by — that the moment may have arrived. ChatGPT bots can now supply the witty repartee, and stunning hot-takes you feel obliged to append to others’ posts for you. You can deploy them as your followers on Twitter.

LinkedIn “Premium” types have been using chatbots to write their posts for some time now. How do we know this? We don’t: this just gives them the benefit of doubt. And if so, why go to the trouble of writing your own dreck in reply, when robots are so much better at it? And once these comment bots truly get going, flaming each other, trolling, hotly disputing unprovable facts about which no educated professional should want an opinion in the first place, then perhaps we humans can step away from LinkedIn together, and let the rich ecosystems of furiously interacting neural networks please themselves.

A fork in the road

There are three paths things can generally go in from a given point and this seems no exception: Good, dreary and bad. JC is a glass half full sort of fellow, but even he struggles to look past the dreary one.

  1. Good: the simulacrum breaks: LinkedIn fizzles into irrelevance, following Facebook and Twitter to the same place MySpace, Second Life and Geocities have already gone. Techbros are pelted with cabbage in the street. Humans kick their online habits and get back out to the healthy offline pursuits of playing IRL bingo, hobby carpentry, community dancing and arguing with Christians in the street.
  2. Dreary: The entropic one, wherein the already tawdry quality of online discourse deteriorates so much that humans depart the playing field altogether leaving the entire social internet to large language models. perhaps this is what Nick Bostrom meant by the [[simulation hypothesis[]: not so much that artificial intelligence crowds human intelligence out, but that the kind of human intelligence that can bear it any longer dwindles to the vanishingly small.
  3. Bad:The dystopian one wherein this is just the impetus the machines need to create some emergent cyber consciousness. We have, by accident, launched SkyNet, only its personality is an amalgam of the slice of humankind that frequents LinkedIn.

In which case, friends, we really are screwed.

See also

References

  1. For thirty years, Grandma Contrarian had the 1981 Royal Wedding taped on video. It was her most prized possession. Not once did any of us watch it.
  2. Regular readers will know the JC is no subscriber to the dismal futurism of Ray Kurzweil or Daniel Susskind.
  3. Happy work-iversary!
  4. Real electric monks, like electric sheep — you know, the ones androids dream of — would be take up space, drain energy and require servicing. Virtual electric monks would not.
  5. If it doesn’t — by no means certain to — then nor does The Man’s, and this phase of our cultural existence will pass on all by itself.