Uber for Nursing: The Algorithmic Wage Discrimination Hiding Behind Surveillance Pricing

Dek: AI’s most quietly devastating use case isn’t taking your job – it’s deciding what your job is worth, one stolen data point at a time.

---

On the July 28 episode of The Weekly Show with Jon Stewart, Cory Doctorow – science-fiction novelist, EFF activist, and the man who gave us the word “enshittification” – made a point that deserves to be pulled out and examined on its own. Everyone is worried about the right AI story: the killer robots, the job-eating chatbot, the paperclip apocalypse. But Doctorow pointed at something AI is actually doing to real workers right now, quietly and profitably. It isn’t replacing radiologists or writing your email. It’s watching you, segmenting you, and deciding – to the dollar – exactly how desperate you are, and therefore how little you can be paid.

It’s called surveillance pricing and algorithmic wage discrimination. And the workers getting squeezed aren’t an easy target. They’re nurses.

“They all call themselves Uber for nursing”

The case that landed hardest came from a report by the Groundwork Collaborative and the Roosevelt Institute. Contract and travel nurses used to be hired through local staffing agencies – the neighborhood body shop that knew you and your work. Not anymore.

“Now, it’s four national apps that compete with each other. They all call themselves Uber for nursing.” – Cory Doctorow, The Weekly Show, 59:20

If the phrase “Uber for nursing” made you feel a cold chill, good. Because the app-ification didn’t stop at convenience. Doctorow explained what those apps do with the data they buy about the very people trying to get work:

“They buy credit card data from data brokers. And the more outstanding credit card debt you have, the lower the wage you’re offered.” – Cory Doctorow, 59:31

Read that again. The app that’s supposed to help you find a nursing shift is buying your credit card data, checking how much you owe, and offering you a lower wage the more desperate you are. Doctorow’s name for it is chillingly precise: a “desperation premium.” Jon Stewart was visibly shaken.

“The company that you’re working for looks to see your level of desperation so that they can lowball you.” – Jon Stewart, 59:42

This is the enshittification endgame in a single sentence: the platform doesn’t just extract value from you – it uses surveillance to figure out the maximum it can take before you have no choice but to say yes.

This is what you need AI for

Doctorow’s most important move was to insist this isn’t a novelty. It’s the point. He invoked Tennessee Ernie Ford’s coal bosses:

“Like me, you have heard of Tennessee Ernie Ford. You know that there are coal bosses who would have done this shit if they could have… But they weren’t going to be able to hire the army of Pinkertons to follow every coal miner around and figure out who was in the most debt, and then get the boiler room full of guys in green eyeshades to mark up the ledgers and change their wages 10 times a minute. This is the thing you need AI for.” – Cory Doctorow, 59:53–1:00:09

That’s the honest sales pitch nobody’s making. The productivity miracle was never about doing your job better. It’s about doing the math on how little you can be squeezed for – at a speed and a scale no human could manage. The coal baron couldn’t afford to have a Pinkerton tail every miner and repriced them ten times a minute. The nursing app can do it to tens of thousands of nurses at once, automatically, before they’ve even finished accepting the shift.

It started with Uber

The nurses’ case is the present. The blueprint was Uber. For 13 years, the ride-hailing giant lost money on almost every ride – “$0.41 on every $1.00 they brought in,” per Doctorow – and burned $31 billion of the Saudi royal family’s money doing it. Then it put the cab companies out of business, became the only game in town, and turned the screw on drivers.

“If you accept a lowball offer, your next offer is lower. And it keeps pushing you down. Now, if you’re picky, it’s higher… whatever it is that’s your side hustle that lets you be picky, eventually, you lose it.” – Cory Doctorow, 57:51–58:08

The driver has twenty seconds to accept or decline. The algorithm learns, in real time, how much each driver can be pushed. And Uber finally got profitable. This is “algorithmic wage discrimination” – a term Doctorow credits to lawyer and scholar Veena Dubal, who has written extensively on it. The wage side has a price-side mirror, surveillance pricing, documented by economist Lindsey Owens in her book Gouged.

“It’s when you use an algorithm to segment an audience based on behavioral surveillance data – I’ve seen what you do. I’ve followed you around and your use of a service. And then you start running experiments on the segments to see what the highest price they’ll accept or the lowest wage they’ll accept.” – Cory Doctorow, 57:19–57:30

The price side: from Mad Men to machine

On the consumer side, the same machinery re-prices the goods you buy in real time. Doctorow pointed out that the old model was almost charming:

“It used to be demographers would do this kind of cute Mad Men style thing… ‘we got the little old ladies who own seven cats who live in Sheboygan.’ And they would go do focus groups.” – Cory Doctorow, 1:00:26–1:00:32

Today, AI “just naively buckets people into groups based on behavioral data, and then automatically runs experiments to see the highest price they will accept.” No focus groups, no human judgment, no shame. Just an optimization loop that knows, from your browsing, your purchases, your card balance, exactly how high it can set the price before you flinch.

The point of AI is to take away your labor power

The most uncomfortable part of the conversation was Doctorow’s refusal to let anyone pretend this is an accident, or a side effect, or a temporary phase. When radiologists – who have real labor power, who are in short supply – tell him they use AI “on our own terms,” he pushes back:

“But the point of AI is to take away your labor power. Right? It’s to find ways to erode your wages and segment you.” – Cory Doctorow, 1:04:38–1:04:49

And he tied it directly to the bubble economics we covered in our piece on AI’s broken unit economics. There’s a reason AI companies can’t make money: their whole pitch to capital is that you’ll eventually be replaced, so the billions in capex will eventually pay off. But that pitch has a very concrete near-term use:

“Pauperize all the cab drivers in the country. Do it again to all the nurses. Do it to all the poor commercial illustrators… and you do not touch the sides of the operating expenditure bills of the AI companies.” – Cory Doctorow, 1:01:49–1:01:58

In other words: the easy targets – the workers who are already paid the least – won’t even dent the AI balance sheet. The whole point is to come for everyone, starting with whoever has the least leverage. Jon Stewart summed it up as the platform finding “another frontier” – “the vast expanse of the ocean or space that they haven’t explored yet… taking what they’ve tried to do in the real world and laying it into the virtual world in that same exploitative manner.” (1:04:49–1:05:02)

Why this matters more than the killer robots

None of this makes the headlines. It’s not dramatic, it doesn’t sing, and there’s no robot with glowing red eyes. But it is, arguably, the most consequential thing AI is actually doing to working people today – as opposed to the things it might do to us someday if the hype is to be believed.

The enshittification pattern is the same every time: bring people in with something good, become the only game in town, then use every bit of power – and every bit of data – to extract more. The new twist is that the extraction is now individualized. You don’t get the same price as your neighbor, or the same wage as your coworker, because the machine has already decided, from your data, exactly how much you can be made to accept.

Nobody votes for this. Nobody goes into a nursing job saying, “please, algorithmically determine my desperation and price my labor accordingly.” As Doctorow noted on the same episode, when Apple gave iPhone users a one-click opt-out from Facebook’s tracking, 96% of people took it – while Apple quietly ran its own opt-out-free ad brokerage in parallel. The surveillance isn’t wanted. It’s simply profitable, because it’s legal, and because there’s been no federal consumer privacy law worth the name since 1988.

“Every other form of privacy invasion is legal in America, right? They spy on you because they can.” – Cory Doctorow, 1:12:46

The fix is boring, and that’s the problem

Doctorow didn’t pretend there’s a clever technical escape. The remedy isn’t a better app or a VPN. It’s the unfashionable stuff: stronger labor protections, antitrust enforcement, and an actual privacy law. The technology only works because the people who could stop it – the same people who let the last 45 years of runaway consolidation happen – choose not to. As he put it in the episode’s closing, the tools to fix this aren’t secret and they aren’t new:

“There are people alive today who learned their trust busting from the people who broke up Rockefeller’s empire. Like, this is- like, they did it. We can do it. They weren’t smarter than us.” – Cory Doctorow, 1:18:31–1:18:35

For more on the economics behind the AI bubble, see the case on AI’s broken unit economics. For the broader frame, the piece on centaurs and reverse centaurs. And for the price side of this same machinery, see The Same Cart, Charged Differently: How Instacart Turned Groceries Into a Surveillance-Pricing Lab.

---

This is a real case, documented by Cory Doctorow on The Weekly Show with Jon Stewart. The excerpt below covers how gig-nursing apps buy data brokers’ credit-card records and price each nurse’s wage by their desperation.

Watch the full episode (The Weekly Show with Jon Stewart) – the nurse wage-discrimination section starts at 59:16.

---

Sources: The Weekly Show with Jon Stewart, episode aired July 28, featuring Cory Doctorow. All timestamps reference that episode. Related work: Veena Dubal (algorithmic wage discrimination), Lindsey Owens, Gouged (surveillance pricing), Groundwork Collaborative & Roosevelt Institute (nursing report).