Pluralistic: A Pascal's Wager for AI Doomers (16 Apr 2026)


Today's links


A killer 1940s robot zapping two large domes with eye-lasers; trapped under the domes are two children, taken from 1910s photos of child laborers; one, a little girl in a straw hat, is holding two heavy buckets. The other, a newsie with a shoulder bag, is picking his nose. The background is the collapsing pillars seen in Dore's engraving of The Death of Solomon.

A Pascal's Wager for AI Doomers (permalink)

Lest anyone accuse me of bargaining in bad faith here, let me start with this admission: I don't think AI is intelligent; nor do I think that the current (admittedly impressive) statistical techniques will lead to intelligence. I think worrying about what we'll do if AI becomes intelligent is at best a distraction and at worst a cynical marketing ploy:

https://locusmag.com/feature/cory-doctorow-full-employment/

Now, that said: among some of the "AI doomers," I recognize kindred spirits. I, too, worry about technologies controlled by corporations that have grown so powerful that they defy regulation. I worry about how those technologies are used against us, and about how the corporations that make them are fusing with authoritarian states to create a totalitarian nightmare. I worry that technology is used to spy on and immiserate workers.

I just don't think we need AI to do those things. I think we should already be worried about those things.

Last week, I had a version of this discussion in front of several hundred people at the Bronfman Lecture in Montreal, where I appeared with Astra Taylor and Yoshua Bengio (co-winner of the Turing Prize for his work creating the "deep learning" techniques powering today's AI surge), on a panel moderated by CBC Ideas host Nahlah Ayed:

https://www.eventbrite.ca/e/artificial-intelligence-the-ultimate-disrupter-tickets-1982706623885

It's safe to say that Bengio and I mostly disagree about AI. He's running an initiative called "Lawzero," whose goal is to create an international AI consortium that produces AI as a "digital public good" that is designed to be open, auditable, transparent and safe:

http://lawzero.org

Bengio said he'd started Lawzero because he was convinced that AI was going to get a lot more powerful, and, in the absence of some public-spirited version of AI, we would be subject to all kinds of manipulation and surveillance, and that the resulting chaos would present a civilizational risk.

Now, as I've stated (and as I said onstage) I am not worried about any of this. I am worried about AI, though. I'm worried a fast-talking AI salesman will convince your boss to fire you and replace you with an AI that can't do your job (the salesman will be pushing on an open door, since if there's one thing bosses hate, it's paying workers).

I'm worried that the seven companies that comprise 35% of the S&P 500 are headed for bankruptcy, as soon as someone makes them stop passing around the same $100b IOU while pretending it's in all their bank accounts at once. I'm worried that when that happens, the chatbots that badly do the jobs of the people who were fired because of the AI salesman will go away, and nothing and no one will do those jobs. I'm worried that the chaos caused by vaporizing a third of the stock market will lead to austerity and thence to fascism:

https://pluralistic.net/2026/04/13/always-great/#our-nhs

I worry that the workers who did those jobs will be scattered to the four winds, retrained or "discouraged" or retired, and that the priceless process knowledge they developed over generations will be wiped out and we will have to rebuild it amidst the economic and political chaos of the burst AI bubble:

https://pluralistic.net/2026/04/08/process-knowledge-vs-bosses/#wash-dishes-cut-wood

In short, I worry that AI is the asbestos we're shoveling into our civilization's walls, and our descendants will be digging it out for generations:

https://pluralistic.net/2026/01/06/1000x-liability/#graceful-failure-modes

But Bengio disagrees. He's very smart, and very accomplished, and he's very certain that AI is about to become "superhuman" and do horrible things to us if we don't get a handle on it. Several times at our events, he insisted that the existence of this possibility made it wildly irresponsible not to take measures to mitigate this risk.

Though I didn't say so at the time, this struck me as an AI-inflected version of Pascal's wager:

A rational person should adopt a lifestyle consistent with the existence of God and should strive to believe in God… if God does not exist, the believer incurs only finite losses, potentially sacrificing certain pleasures and luxuries; if God does exist, the believer stands to gain immeasurably, as represented for example by an eternity in Heaven in Abrahamic tradition, while simultaneously avoiding boundless losses associated with an eternity in Hell.

https://en.wikipedia.org/wiki/Pascal%27s_wager

Smarter people than me have been poking holes in Pascal's wager for more than 350 years. But when it comes to this modern Pascal's AI Wager, I have my own objection: how do you know when you've lost?

As of this moment, the human race has lit more than $1.4t on fire to immanentize this eschaton, and it remains stubbornly disimmanentized. How much more do we need to spend before we're certain that god isn't lurking in the word-guessing program? Sam Altman says it'll take another $2-3t – call it six months' worth of all US federal spending. If we do that and we still haven't met god, are we done? Can we call it a day?

Not according to Elon Musk. Musk says we need to deconstruct the solar system and build a Dyson sphere out of all the planets to completely encase the sun, so we can harvest every photon it emits to power our word-guessing programs:

https://www.pcmag.com/news/elons-next-big-swing-dyson-sphere-satellites-that-harness-the-suns-power

So let's say we do that and we still haven't met god – are we done? I don't see why we would be. After all, Musk's contention isn't that our sun emits one eschaton's worth of immanentizing particles. Musk just thinks that we need a lot of these sunbeams to coax god into our plane of existence. If one sun won't do it, perhaps two? Or two hundred? Or two thousand? Once we've committed the entire human species to this god-bothering project to the extent of putting two kilosuns into harness, wouldn't we be nuts to stop there? What if god is lurking in the two thousand and first sun? Making god out of algorithms is like spelling "banana" – easy to start, hard to stop.

But as Bengio and I got into it together on stage at the Montreal Centre, it occurred to me that maybe there was some common ground between us. After all, when someone starts talking about "humane technology" that respects our privacy and works for people rather than their bosses, my ears grow points. Throw in the phrase "international digital public goods" and you've got my undivided attention.

Because there's a sense in which Bengio and I are worried about exactly the same thing. I'm terrified that our planet has been colonized by artificial lifeforms that we constructed, but which have slipped our control. I'm terrified that these lifeforms corrupt our knowledge-creation process, making it impossible for us to know what's true and what isn't. I'm terrified that these lifeforms have conquered our apparatus of state – our legislatures, agencies and courts – and so that these public bodies work against the public and for our colonizing alien overlords.

The difference is, the artificial lifeforms that worry me aren't hypothetical – they're here today, amongst us, endangering the very survival of our species. These artificial lifeforms are called "limited liability corporations" and they are a concrete, imminent risk to the human race:

https://pluralistic.net/2026/04/15/artificial-lifeforms/#moral-consideration

What's more, challenging these artificial lifeforms will require us to build massive, "international, digital public goods": a post-American internet of free/open, auditable, transparent, enshittification-resistant platforms and firmware for every purpose and device currently in service:

https://pluralistic.net/2026/01/01/39c3/#the-new-coalition

And even after we've built that massive, international, digital public good, we'll still face the challenge of migrating all of our systems and loved ones out of the enshitternet of defective, spying, controlling American tech exports:

https://pluralistic.net/2026/01/30/zucksauce/#gandersauce

Every moment that we remain stuck in the enshitternet is a moment of existential risk. At the click of a mouse, Trump could order John Deere to switch off all the tractors in your country:

https://pluralistic.net/2022/05/08/about-those-kill-switched-ukrainian-tractors/

He doesn't need tanks to steal Greenland. He can just shut off Denmark's access to American platforms like Office365, iOS and Android and brick the whole damned country. It would be another Strait of Hormuz, but instead of oil and fertilizer, he'd control the flow of Lego, Ozempic and deliciously strong black licorice:

https://pluralistic.net/2026/01/29/post-american-canada/#ottawa

These aren't risks that could develop in the future. They're the risks we're confronted with today and frankly, they're fucking terrifying.

So here's my side-bet on Pascal's Wager. If you think we need to build "international digital public goods" to head off the future risk of a colonizing, remorseless, malevolent artificial lifeform, then let us agree that the prototype for that project is the "international digital public goods" we need right now to usher in the post-American internet and save ourselves from the colonizing, remorseless, malevolent artificial lifeforms that have already got their blood-funnels jammed down our throats.

Once we defeat those alien invaders, we may find that all the people who are trying to summon the evil god have lost the wherewithal to do so, and your crisis will have been averted. But if that's not the case and the evil god still looms on our horizon, then I will make it my business to help you mobilize the legions of skilled international digital public goods producers who are still flush from their victory over the limited liability corporation, and together, we will fight the evil god you swear is in our future.

I think that's a pretty solid offer.

Hey look at this (permalink)


A shelf of leatherbound history books with a gilt-stamped series title, 'The World's Famous Events.'

Object permanence (permalink)

#25yrsago Every pirate ebook on the internet https://web.archive.org/web/20010724030402/https://citizen513.cjb.net/

#20yrsago Retired generals diss Donald Rumsfeld https://nielsenhayden.com/makinglight/archives/007432.html#007432

#20yrsago How to break HDCP https://blog.citp.princeton.edu/2006/04/14/making-and-breaking-hdcp-handshakes/

#20yrsago How Sunā€™s ā€œopen DRMā€ dooms them and all they touch https://memex.craphound.com/2006/04/14/how-suns-open-drm-dooms-them-and-all-they-touch/

#20yrsago Benkler's "Wealth of Networks" http://www.congo-education.net/wealth-of-networks/

#15yrsago Scientific managementā€™s unscientific grounding: the Management Myth https://web.archive.org/web/20120823212827/https://www.theatlantic.com/magazine/archive/2006/06/the-management-myth/304883/

#15yrsago 216 ā€œuntranslatableā€ emotional words from non-English languages https://www.drtimlomas.com/lexicography/cm4mi/lexicography#!lexicography/cm4mi

#10yrsago New York public employees union will vote on pulling out of hedge funds https://web.archive.org/web/20160414230326/https://www.bloomberg.com/news/articles/2016-04-13/nyc-pension-weighs-liquidating-1-5-billion-hedge-fund-portfolio

#10yrsago Panamaā€™s public prosecutor says he canā€™t find any evidence of Mossack-Fonsecaā€™s lawbreaking https://web.archive.org/web/20160419165306/https://www.thejournal.ie/mossack-fonseca-prosecution-2714795-Apr2016/?utm_source=twitter_self

#10yrsago Bernie Sanders responds to CEOs of Verizon and GE: ā€œI welcome their contemptā€ https://web.archive.org/web/20160415165051/https://www.businessinsider.com/bernie-sanders-verizon-contempt-2016-4

#10yrsago Letā€™s Encrypt is actually encrypting the whole Web https://www.wired.com/2016/04/scheme-encrypt-entire-web-actually-working/

#10yrsago City of San Francisco tells man he canā€™t live in wooden box in friendā€™s living room https://www.theguardian.com/us-news/2016/apr/13/san-francisco-new-home-rented-box-illegal?CMP=tmb_gu

#10yrsago How the UKā€™s biggest pharmacy chain went from family-run public service to debt-laden hedge-fund disaster https://www.theguardian.com/news/2016/apr/13/how-boots-went-rogue

#10yrsago Ohio newspaper chain owner says his papers donā€™t publish articles about LGBTQ people https://ideatrash.net/2016/04/the-owner-of-four-town-papers-in-ohio.html

#10yrsago How British journalists talk about people theyā€™re not allowed to talk about https://web.archive.org/web/20160414152933/https://popbitch.com/home/2016/03/31/up-the-injunction/

#10yrsago Brussels terrorists kept their plans in an unencrypted folder called ā€œTARGETā€ https://www.techdirt.com/2016/04/14/brussels-terrorist-laptop-included-details-planned-attack-unencrypted-folder-titled-target/

#10yrsago Ron Wyden vows to filibuster anti-cryptography bill https://www.techdirt.com/2016/04/14/burr-feinstein-officially-release-anti-encryption-bill-as-wyden-promises-to-filibuster-it/

#10yrsago Paramount wants to kill a fan-film by claiming copyright on the Klingon language https://torrentfreak.com/paramount-we-do-own-the-klingon-language-and-warships-160414/

#5yrsago Murder Offsets https://pluralistic.net/2021/04/14/for-sale-green-indulgences/#killer-analogy

#5yrsago The FCC wants your broadband measurements https://pluralistic.net/2021/04/14/for-sale-green-indulgences/#fly-my-pretties

#1yrago Machina economicus https://pluralistic.net/2025/04/14/timmy-share/#a-superior-moral-justification-for-selfishness

Upcoming appearances (permalink)

A photo of me onstage, giving a speech, pounding the podium.


A screenshot of me at my desk, doing a livecast.

Recent appearances (permalink)


A grid of my books with Will Stahle covers..

Latest books (permalink)


A cardboard book box with the Macmillan logo.

Upcoming books (permalink)

  • "The Reverse-Centaur's Guide to AI," a short book about being a better AI critic, Farrar, Straus and Giroux, June 2026 (https://us.macmillan.com/books/9780374621568/thereversecentaursguidetolifeafterai/)

  • "Enshittification, Why Everything Suddenly Got Worse and What to Do About It" (the graphic novel), Firstsecond, 2026

  • "The Post-American Internet," a geopolitical sequel of sorts to Enshittification, Farrar, Straus and Giroux, 2027

  • "Unauthorized Bread": a middle-grades graphic novel adapted from my novella about refugees, toasters and DRM, FirstSecond, 2027

  • "The Memex Method," Farrar, Straus, Giroux, 2027


Colophon (permalink)

Today's top sources:

Currently writing: "The Post-American Internet," a sequel to "Enshittification," about the better world the rest of us get to have now that Trump has torched America. Third draft completed. Submitted to editor.

  • "The Reverse Centaur's Guide to AI," a short book for Farrar, Straus and Giroux about being an effective AI critic. LEGAL REVIEW AND COPYEDIT COMPLETE.

  • "The Post-American Internet," a short book about internet policy in the age of Trumpism. PLANNING.

  • A Little Brother short story about DIY insulin PLANNING

This work – excluding any serialized fiction – is licensed under a Creative Commons Attribution 4.0 license. That means you can use it any way you like, including commercially, provided that you attribute it to me, Cory Doctorow, and include a link to pluralistic.net.

https://creativecommons.org/licenses/by/4.0/

Quotations and images are not included in this license; they are included either under a limitation or exception to copyright, or on the basis of a separate license. Please exercise caution.

How to get Pluralistic:

Blog (no ads, tracking, or data-collection):

Pluralistic.net

Newsletter (no ads, tracking, or data-collection):

https://pluralistic.net/plura-list

Mastodon (no ads, tracking, or data-collection):

https://mamot.fr/@pluralistic

Bluesky (no ads, possible tracking and data-collection):

https://bsky.app/profile/doctorow.pluralistic.net

Medium (no ads, paywalled):

https://doctorow.medium.com/

Tumblr (mass-scale, unrestricted, third-party surveillance and advertising):

https://mostlysignssomeportents.tumblr.com/tagged/pluralistic

"When life gives you SARS, you make sarsaparilla" -Joey "Accordion Guy" DeVilla

READ CAREFULLY: By reading this, you agree, on behalf of your employer, to release me from all obligations and waivers arising from any and all NON-NEGOTIATED agreements, licenses, terms-of-service, shrinkwrap, clickwrap, browsewrap, confidentiality, non-disclosure, non-compete and acceptable use policies ("BOGUS AGREEMENTS") that I have entered into with your employer, its partners, licensors, agents and assigns, in perpetuity, without prejudice to my ongoing rights and privileges. You further represent that you have the authority to release me from any BOGUS AGREEMENTS on behalf of your employer.

ISSN: 3066-764X

Join the Discussion at ChinWag