Pluralistic: Bullies want you to think they're on your side (13 Mar 2024)

Originally published at: Pluralistic: Bullies want you to think they’re on your side (13 Mar 2024) – Pluralistic: Daily links from Cory Doctorow

Today's links

A fortress with a drawbridge. On the drawbridge stands a beckoning military figure with a demon's head. Through the portcullis gate we see a mousetrap against a backdrop of books.

Bullies want you to think they're on your side (permalink)

Bruce Schneier coined the term "feudal security" to describe Big Tech's offer: "move into my fortress – lock yourself into my technology – and I will keep you safe from all the marauders roaming the land":

But here's the thing about trusting a warlord when he tells you that the fortress's walls are there to keep the bad guys out: those walls also keep you in. Sure, Apple will use its control over Ios to stop Facebook from spying on you, but when Apple spies on you, no one can help you, because Apple exercises total control over all Ios programs, including any that would stop Apple from nonconsensually harvesting your data and selling access to it:

It's a tried-and-true bullying tactic: convince your victim that only you can keep them safe so they surrender their agency to you, so the victim comes under your power and can't escape your cruelty and exploitation. The focus on external threats is key: so long as the victim is more afraid of the dangers beyond the bully's cage than they are of the bully, they can be lured deeper and deeper until the cage-door slams shut.

Think of how the capital class talks about labor and immigration. For 40 years, workers' wages have stagnated, even as workers grew more productive and GDP grew and grew. In other words, workers held up their end of the bargain – creating value for the companies that employed them – but their bosses reneged on the part of the deal where they got a fair share of that value.

Historically, there are two ways that workers get a fair deal: either they get bargaining leverage because there aren't enough workers to go around, or they form a union and get bargaining leverage through solidarity. The former never works for very long: if wages are high because of worker scarcity, other workers will move or retrain until wages start to fall again. The only way to keep wages high is for workers to unionize and meet their bosses' concentrated power of capital with the concentrated power of a union:

Bosses know that the reason that workers' wages stagnated is because they smashed unions, but they told workers that it was because "immigrants stole their jobs." This set the stage for decades of increasingly authoritarian measures ("build the wall") that weaken the rights of all people – but especially brown people. They say they're making a fortress to keep others out, but really they're making a prison to keep us in.

Tech is especially skilled at playing this fortress/prison gambit. 20 years ago, Amazon claimed that publishers who used "digital rights management" for Kindle ebooks and Audible audiobooks would be protected from their real enemy – pirates. But the real threat to publishers was Amazon, not pirates. Amazon was the entity hellbent on extracting huge discounts from publishers, even – especially – if it drove those publishers into bankruptcy:

By locking their titles to Amazon's platform with DRM, publishers surrendered their bargaining leverage with the tech giant. Every time a publisher sells a DRM-"protected" book to a customer, they add to the cost that customer will have to bear if they ever quit Amazon, because only Amazon-approved apps can read DRM-encumbered files sold by Amazon:

Jeff Bezos is a canonical protector-turned-jailer. By tricking publishers into permanently shackling their products to technology he controlled, he was able to shift billions in value from their side of the ledger to his own – and he did it by telling them he was keeping publishers safe!

Not that entertainment companies are always on the side of the angels. Today, entertainment companies – publishers, stock photo companies, etc – are chasing "AI" companies through the courts, seeking to eliminate the fair use status of making transient copies of works, performing mathematical analysis on those copies, and turning the results of that analysis into software:

The thing is, it's not really controversial among copyright experts to say that these activities are fair use. Scraping isn't stealing – not legally and not ethically. Scraping is good, actually:

Which is not to say that artists don't have a legit – and serious – beef with AI companies. Their pitch is: "train our products on your workers' skilled output, fire your workers, replace them with our products." That's a monstrous proposition – and that's before we get to the part where their products aren't anywhere near good enough to do your job, but their salesmen are absolutely good enough to convince your boss to fire you and replace you with an AI model that totally fails to do your job.

But while the corporations that pay artists may hate the idea of AI companies scraping their archives, they hate paying the creative workers that fill those archives every bit as much. If we end up changing the law to make scraping, mathematical analysis, and publication of that analysis without permission illegal, it won't do anything to stop the media corporations suing AI companies from building models with the intent of firing workers and replacing them with software.

More than a year ago, I predicted that the very same companies that are suing over AI training would happily sell their data to AI companies. They don't want "no AI" – they want "AI that they control":

Today, Apple is offering tens of millions to media companies for access to creative works in order to train models. These media companies require overbroad, one-sided contracts with creative workers that give them the power to sell their work to Apple:

Media companies aren't on their workers' side in just the same way that Amazon wasn't on publishers' side. The promise of a new, broader copyright that bans scraping, mathematical analysis and writing code is like the promise of DRM: it's a prison pretending to be a fortress.

Because if that new copyright law controlling training is created, media companies will immediately amend those one-sided, non-negotiable, take-it-or-leave-it contracts to demand that every creative worker sign over that new right to them. Hell, they're already doing this in anticipation of that change:

Remember, AI companies make money by selling models to companies hoping to fire their workers. But for that to work, companies have to plan on making money by firing their workers and replacing them with AI. If no company was interested in this proposition, we wouldn't have to worry about AI. Capital and labor are not on the same side. Creative workers don't have the same problem with AI as the media companies that want to drive down their wages, and the "solution" that media corporations favor isn't going to do a damned thing for workers – but it'll do plenty to them.

The media companies providing training data for Apple's "ethical" AI model will use it in exactly the same way that the media companies who're slavering over the "nonethical" AI models want to use them.

Meanwhile, every time a worker – or their ally – is tricked into calling scraping "stealing," they do the bosses' job for them. And the bosses – being remorseless seekers of return to capital – don't care about the collateral damage in their war on scraping.

Scraping is how we find out if Facebook is making money by selling paid political disinformation:

Scraping is how we preserve the web:

Scraping is how journalists rescue their work from media companies after private equity owners shut them down and delete their archives:

Scraping is how we'll get new, better search-engines that filter out the botshit, scams and SEO nonsense:

Media bosses also don't care about the collateral damage in their war on mathematical analysis – right to make temporary copies of works and subject them to scrutiny. They don't care about extinguishing the entire exciting, useful discipline of computational linguistics:

They don't care about inadvertently banning the use of language models to analyze police records for the Innocence Project New Orleans:

Or the records of killings during the Colombian civil war, to produce evidence that the vast majority of these killings were carried out by CIA-backed right-wing militias:

Back in the Napster Wars, when record companies sued 19,000 American college kids, single moms, and dead people (!), a RIAA spokesman notoriously said, "When you fish with a net, you are going to catch a few dolphins":

If this Copyright Wars 2.0: AI Edition ever gets going in earnest – if we manage to snuff out the right to scrape and the right to analyze without permission – we won't stop media companies from selling creative workers' output to AI companies. We won't stop them from using AI models to suppress our wages. We will kill all the beneficial uses of scraping and analyzing.

In other words, we'll create a tuna net that catches nothing but dolphins.

How do we make workers whole? How do we protect their interests? Remember, there are only two ways that workers have ever gotten a fair shake from capital. The first is to exploit a temporary and doomed shortage of labor for short-lived bargaining advantage. The second is to form a union and force our bosses to give us what we deserve. The only successful campaign against AI – the only campaign that actually advantaged workers, not their bosses – was the Hollywood strikes last summer.

What's more, in this fight – finally, for once! – workers have the law on our side. The US Copyright Office has repeatedly confirmed (and courts have upheld) that the output from AIs isn't eligible for copyright, meaning that if our bosses ever manage to replace us with algorithms, they can't stop their rivals from scraping the output of those bots and selling it in direct competition with them:

In other words, adding copyright to AI will make it easier for our bosses to fire us and replace us with botshit.

(Image: Skelanard, CC BY-SA 4.0)

Hey look at this (permalink)

A Wayback Machine banner.

This day in history (permalink)

#20yrsago We can use the power outlets at SXSW now!

#10yrsago RIP, Tony Benn

#10yrsago Tony Benn, secret mounter of illegal Parliamentary plaques

#10yrsago Vi Hart’s updated poop-on-Pi video<?a.

#10yrsago Zuckerberg phones Obama to complain about NSA spying

#5yrsago Walt Disney’s Frozen Head: a science fiction movie secretly shot at Walt Disney World

#5yrsago Facebook and Big Tech are monopsonies, even when they’re not monopolies

#5yrsago A detailed analysis of American ER bills reveals rampant, impossible-to-avoid price-gouging

Upcoming appearances (permalink)

A photo of me onstage, giving a speech, holding a mic.

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)

  • Picks and Shovels: a sequel to "Red Team Blues," about the heroic era of the PC, Tor Books, February 2025
  • Unauthorized Bread: a graphic novel adapted from my novella about refugees, toasters and DRM, FirstSecond, 2025

Colophon (permalink)

Today's top sources:

Currently writing:

  • A Little Brother short story about DIY insulin PLANNING
  • Picks and Shovels, a Martin Hench noir thriller about the heroic era of the PC. FORTHCOMING TOR BOOKS JAN 2025

  • Vigilant, Little Brother short story about remote invigilation. FORTHCOMING ON TOR.COM

  • Spill, a Little Brother short story about pipeline protests. FORTHCOMING ON TOR.COM

Latest podcast: The Majority of Censorship is Self-Censorship

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

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):

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

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

Medium (no ads, paywalled):

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

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

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

1 Like

Regarding scraping, I was shocked to see that OCLC, the nonprofit company that runs WorldCat, is suing an online archive (Anna’s Archive) for scraping their data. First of all, OCLC doesn’t create the data it sells. Catalogers in libraries across the world create that data and consider it a commons. Especially since OCLC is A NONPROFIT. Second of all, with likely many different entities scraping WorldCat every day, it seems so weird to target the online archive that was grabbing the data just to do an analysis on what they were missing. With all the rampant misuse of scraped data for AI out there, it just seems like a really misguided lawsuit. If they win they will set precedent that will be detrimental to the whole community of librarians and archivists that OCLC currently purportedly serves. More info : Lawsuit Accuses Anna's Archive of Hacking WorldCat, Stealing 2.2 TB Data * TorrentFreak

1 Like

This topic was automatically closed after 15 days. New replies are no longer allowed.