The Worst Case Scenario For AI Is Already Here

The first malevolent AI was the VoC in the 1600s. De gepersonifieerde VOC ontvangt geschenken, Jan Caspar Philips, 1730. The personified VOC receives ‘gifts’

If you expand your definition of AI to match the legal definition, it’s already here. Corporations have had legal personhood for centuries and used that to run riot over the world. Just because these artificial beings have human components we don’t call them AI, but we’re the ones getting squashed. Corporate AI is AI, and we ignore that established legal and political fact at our peril. Ignorant pride won’t protect us from what’s going on.

You can call corporations cyborgs if you want, ie a combination of man and machine. Whether you call the cyborgs or AI, they are artificial constructs that can act upon the world with agency and are already acting up. As I wrote earlier:

When Mitt Romney said “corporations are people, my friend” people laughed at him. But he was right. Corporations are legal persons. This is not an analogy or a metaphor but a legal fact. Corporate personhood has been accepted since ancient India, through Rome, to the present day.

Corporations can buy property, enter contracts, and are treated generally the same as ‘natural’ persons in the eyes of the law. They even have the right to free speech, including influencing elections. Corporations are artificial persons. Therefore (since we assume people are intelligent), they are artificially intelligent. Corporations are AI.

Corporations are already fully legal artificial beings who are already fucking shit up. This has been legally established for centuries and politically evident for decades at least (for white people) and ever since colonialism for the rest of us, which was corporations viciously preying on us.

Hence the worst case for AI is already here and has been for centuries for most of the world. But because they encountered a convincing chatbot, commentators are vainly writing think pieces about how to avoid the worst case. My brother in capitalism, we’re already fucked. As some dude writes in the New Yorker (I only read until the paywall kicked in):

In the worst-case scenario envisioned by these thinkers, uncontrollable A.I.s could infiltrate every aspect of our technological lives, disrupting or redirecting our infrastructure, financial systems, communications, and more. Fake people, now endowed with superhuman cunning, might persuade us to vote for measures and invest in concerns that fortify their standing, and susceptible individuals or factions could overthrow governments or terrorize populations. (New Yorker 🤮)

Honestly, these people aren’t thinking hard enough. What the writer describes here is what corporations are already doing. Infiltrating our technological lives (Amazon, literally everyone), messing up infrastructure for profit (Enron, fossil fuel companies, all the privatized train wrecks), fucking up the financial world (BlackRock, everyone), corrupting communications (Facebook, et al). Persuading us to vote for measures we don’t want and fortify their standing? This is literally being a corporation 101. Lobbying, bribing politicians, and couping governments they don’t like? This is old hat by now. Just because corporations act through humans parts doesn’t make it less of an Artificial Intelligence. Missing that point is a failure of our intelligence, not theirs.

I will return again and again to the point that corporations are already legal persons, with most of the rights of humans, even more rights that humans don’t have (like getting away with murder), and basically none of the responsibility (limited liability FTW). Whatever threshold we think AI is going to have to clear to ‘be accepted’ has already been leapt over long, long, ago while we sit around debating semantics. The first AI was the Dutch East India Company in the 1600s and it literally enslaved people and stripped the Earth. And the corporations haven’t stopped. This is not something ChatGPT is going to do to us in the future. It’s already been done.

In this sense ‘talking’ AI is a bit of a party trick. Corporate AI has been talking through legal documents and corporate statements for centuries. You could say that it’s been humans doing and directing this, but are we? Bacteria digests most of my food, but I’m still given agency for my farts. Any living being beyond the tiniest contains multiple living beings, we are more than the sum of our parts.

Humans are vital cogs in these machine, but they direct the thing not much more than cogs do. Even CEOs, who ostensibly ‘control’ individual firms, are programmatically removed by boards if they don’t maximize the life-cycle of the corporation (ie, profits). Across the ecosystem of the market, machines that don’t serve the needs of the AI species will simply be removed and tossed out in favor of those that do. What we have in corporations are globe-spanning AI which contain human parts, which are swapped out as soon as possible.

Does this qualify as AI? What is the point have having a philosophical debate about something that’s already legally and politically established, and which is already eating up your time and the Earth you live on. This is like debating whether a lion is a carnivore while you’re in its mouth. Legally, the status of corporations as artificial persons is well established, and politically they’re already running the joint.

Hence all the fears about AI are not about some distant point in the future. The worst case is already here, and has been for centuries. Just look around and the force and power of corporations in your life already. Look at how much power corporations already have over human life, the land, every other species, law, politics, and everything we hold dear. You don’t need a chatbot to tell you this. Just look around at your own experience using just the legal definition of artificial personhood. The worst case for AI ain’t in the future. It’s already here.

I’ve written a lot about corporations as AI here.