"I spoke to a playwright making $10,000 a month, a multitalented chemist who at various points found gigs demonstrating poker and singing for AI."
@joshdzieza.bsky.social for @theverge.com: www.theverge.com/cs/features/...
AI companies are hiring white-collar professionals to produce the training data needed to automate their jobs, @joshdzieza.bsky.social reports. This massive harvesting of human expertise is enabled by “an arrangement in which employers can turn labor on and off like a tap.” nymag.com/intelligence...
Workers began posting complaints on Mercor’s subreddit, only to have their posts quickly deleted by the Mercor representatives who moderate it. In response, two unsanctioned Mercor subreddits were created, where workers could freely express such sentiments as “CHILDREN RUN THIS COMPANY, THEY WILL SOON HAVE THEIR DAY OF RECKONING.”
This reporting by @joshdzieza.bsky.social is harrowing. The arbitrary, precarious, and piecemeal work done by professionals to train generative AI systems.
The data work contractor, Mercor, founded by 19-year olds, pull contracts and change working conditions at the drop of a hat.
The workers at the bottom of this supply chain exist in a state of extreme precarity and maximum competitive frenzy — especially because their strict confidentiality agreements make it impossible for them to establish any kind of seniority or relationship that might outlast a particular project. “The power is all on one side because they can’t talk about it,” says Matthew McMullen, a strategy and operations executive who has worked in the industry since the self-driving-car boom in the mid-2010s. “The labs benefit from you not being able to leverage your experience in the market, and this silence is like their pricing power. The silence is their ability to extract mass information from people without giving them the power to object or to unionize or to make companies themselves. As long as they can’t prove what they’ve done, these raters can’t demand what they’re worth. The only way that people can demand things is by showing their ability to step up, to take on more work. The only power that they have is to keep going, to get back in line.”
The AI data supply chain is deliberately opaque. Projects are code named. Workers rarely know who they're producing data for -- it's always just "the client." NDAs bar them from describing their work at any level of detail. www.theverge.com/cs/features/...
The intermittent work, productivity monitoring, and surplus of AI-vetted contractors means that workers are effectively always on call, often work overtime, and are paid for only the minutes they are typing. www.theverge.com/cs/features/...
Data work is extremely precarious. Projects start and stop without warning. Workers have their activity tracked and time deemed "unproductive" can be deducted from their pay. Automatic, unexplained firing is common. www.theverge.com/cs/features/...
Cannot recommend taking the time to read this article more strongly. Truly compelling reporting that does the best work I have seen in a long time depicting the jobs that AI is creating: precarious data works. Huge congrats to @joshdzieza.bsky.social.
One type of project had teams of HR managers, lawyers, or bankers role-play their jobs inside a simulated corporate environment, producing fictional slide decks, financial analyses, and meeting notes with which to test train AI models. www.theverge.com/cs/features/...
AI labs are paying billions of dollars for data on practically any job you can think of: consultants, chefs, private investigators, graphic designers, teachers, archivists, wildlife conservation scientists. www.theverge.com/cs/features/...
AI companies are paying screenwriters, lawyers, and other white-collar professionals to produce the training data needed to automate their jobs. I spoke with more than 30 workers about conditions inside this fast-growing and extremely secretive new gig economy.
Enjoyed going on Shifting Terrain to talk about AI, training data, and work: podcasts.apple.com/us/podcast/a...
Verge headline: Jimmy Wales on Wikipedia, bias, and the "Gaza genocide" page Photo: Jimmy Wales
Wikipedia is under attack — from accusations of bias, from AI scrapers, from Elon Musk — but the encyclopedia’s founder believes that transparency is the key to survival.
Read @joshdzieza.bsky.social's interview with Jimmy Wales: www.theverge.com/tech/846184/...
AI companies tout their spending on data centers but are quieter about training data, where their spending is also surging. One reason may be that increased spending on task-specific data looks more like the trajectory of a normal technology than imminent AGI. www.theverge.com/cs/features/...
As scaling shows diminishing returns, AI companies are turning to domain-specific training in areas like coding and finance. The billions they're spending on data is reconfiguring everything from staffing agencies to job-interview platforms. www.theverge.com/cs/features/...
It's a new day, and that means a new batch of Verge articles are completely free to read (until tomorow).
First up, from @joshdzieza.bsky.social: The world’s largest encyclopedia became the factual foundation of the web, but now it’s under attack.
www.theverge.com/cs/features/...
The surging demand for specialized training data cuts against the idea of imminent AGI. AGI should generalize, not require bespoke data for every task. www.theverge.com/cs/features/...
Current AI training methods work for math and coding, where success is verifiable. But most domains aren't like that, so companies are paying lawyers, doctors, poets, woodworkers, etc to write super-specific checklists for everything they might do on the job.
AI companies are spending billions hiring humans to produce training data. @haydenfield.bsky.social and I wrote about the explosion in new vendors and what it means for the future of AI development.
Great article on how Wikipedia uses process, often unto tedium, to prevent polarization and fend off ideological attacks. But pay attention to the part about “state capture,” and what happened in Croatia when neo-Nazis took over the admin board.
This long read in The Verge does a remarkable job of describing how Wikipedia's editing community works, the project's strengths and weaknesses, and the threats it faces.
www.theverge.com/cs/features/...
If you haven't read this @joshdzieza.bsky.social piece, I strongly suggest you do. Too much good stuff to single out one quote, but the frame it offers: verification and deliberative decision-making as a bulwark against power, is a vital antidote to the vanities of the current media discourse
just learned about the croatian wikipedia coup of 2010 and also this associated headline which might be the best headline ever written
"Wikipedia is one of the few platforms online where tremendous computing power isn’t being deployed in the service of telling you exactly what you want to hear."
Terrific Verge piece this morning www.theverge.com/cs/features/...
In the US, Wikipedia editors see a similar playbook being deployed. Conservative outlets and influencers have been attacking the site, and last week House Oversight demanded info on alleged efforts to “inject bias” into the encyclopedia, including details on individual editors.
In India, a pro-gov media company is suing over alleged defamation while far-right publications and influencers dox and harass editors, including referring them to police for investigation. It has had a chilling effect on volunteers. www.theverge.com/cs/features/...
Governments, billionaires, influencers, and political groups are increasingly trying to undermine and manipulate Wikipedia. In places where social media, journalism, and academia have been brought under control, the encyclopedia is often the next target.
The most contentious Wikipedia articles are the highest quality, because dueling editors keep adding sources in support of their view. Articles (and editors themselves) become more moderate over time. It's basically the opposite of algorithmic attention-maxxing sites www.theverge.com/cs/features/...
Conflicts on Wikipedia can be extraordinarily protracted. 40K words about capitalization! But bc they hinge on who best follows wiki process, even disagreeing editors are affirming the project's basic principles around sourcing, neutrality, etc www.theverge.com/cs/features/...
