Think of all the things that don’t make it onto your calendar, like making your bed in the morning, cooking yourself dinner, and cleaning up after.
They’re mundane tasks, but ones that must be done nonetheless.
At least now you could be paid to do them — and already, many people are.
Why?
They’re part of a growing workforce of gig workers who, in exchange for cash, are helping to train AI by recording themselves performing everyday tasks, so that eventually physical bots can do it in their place.
- In LA, Instawork has people strap on a phone-mountable headset and wristbands that collect data on their movements as they work, per the Los Angeles Times. Each task, anything from plant watering to dishwashing, must be 2-15 minutes long. Two hours of footage could net a cool ~$80.
- DoorDash is tapping into its network of couriers with Tasks, a standalone app it announced this month, where Dashers can accept assignments — e.g., recording themselves folding clothes or speaking a foreign language — with upfront rates based on complexity and effort, per Bloomberg.
“It’s one of the biggest gig economies that is going to exist in the whole world,” the founder of one such startup told LA Times of the global human data capture market, which is projected to hit ~$17B by 2030.
Everyone’s a gig worker now
Even laid-off white-collar professionals are now training AI to do their jobs, The Wall Street Journal reports, thanks to companies like Scale AI, Mercor, and Surge AI, whose ranks comprise tens or even hundreds of thousands of experts, including VCs, lawyers, astronomers, and writers.
- Some people assume the job postings are scams because the roles pay pretty well — as much as $250 an hour for certain specialists and up to $150 for poets — but that’s because it kinda requires you to sell your soul, or at least, in some cases, the rights to your IP.
- Many feel uneasy about the work, but for job seekers facing months of unemployment, it’s hard to blame them for taking the gigs, even if they might be digging their own career graves.
“I didn’t invent AI and I’m not going to uninvent it,” one laid-off journalist, who now spends 20-30 hours editing AI articles, told WSJ. “If I were to stop doing this, would that stop it? The answer is no.”
Bleak, but fair enough.






Comments