After the birth of her second child, Chelsea Becker took a year-long unpaid leave from her full-time job as a flight attendant. After watching a video on TikTok, she found a side job: training artificial intelligence models for a website called Data Annotation tech.
For a few hours each day, Becker, 33, who lives in Schwenksville, Pennsylvania, sat in front of his laptop and interacted with an ai-powered chatbot. For each hour of work, he was paid between $20 and $40. From December to March, he earned more than $10,000.
The rise of artificial intelligence technology has given a more sophisticated twist to a type of informal work that does not require leaving the home. The growth of large language models, such as the technology powering OpenAI's ChatGPT, has driven the need for trainers like Ms. Becker, who speak fluent English and can produce quality writing.
It's no secret that ai models learn from humans. For years, makers of artificial intelligence systems like Google and OpenAI have relied on low-paid workers, usually contractors employed through other companies, to help computers visually identify subjects. (The New York Times has sued OpenAI and its partner, Microsoft, over allegations of copyright infringement.) They could tag vehicles and pedestrians to be autonomous vehicles or identify images in photographs used to train ai systems.
But as ai technology has become more sophisticated, so has the job of the people who must thoroughly teach it. Yesterday's photo tagger is today's essay writer.
There are typically two types of work for these trainers: supervised learning, where the ai learns from human-generated writing, and reinforcement learning from human feedback, where the chatbot learns from how humans rate its responses.
Companies that specialize in data curation, including San Francisco-based startups Scale ai and Surge ai, hire contractors and sell their training data to larger developers. ai model developers, such as Toronto-based startup Cohere, also recruit internal data annotators.
It is difficult to estimate the total number of these workers, the researchers said. But Scale ai, which hires contractors through its subsidiaries, Remotasks and Outlier, said it was common to see tens of thousands of people working on the platform at any given time.
But as with other types of informal work, the ease of flexible hours comes with its own challenges. Some workers said they never interacted with the administrators behind the recruiting sites and others had been excluded from work without explanation. Researchers have also expressed concern about a lack of standards, as workers generally do not receive training on what are considered appropriate chatbot responses.
To become one of these contractors, workers must pass an assessment, which includes questions such as whether a social media post should be considered hateful and why. Another requires a more creative approach, asking prospective hirers to write a fictional short story about a dancing green octopus, set in Sam Bankman-Fried's FTX offices on November 8, 2022 (that was the day Binance, a competitor of FTX, said it would buy Mr. Bankman-Fried's company before quickly backing out of the deal).
Sometimes companies look for subject matter experts. Scale that ai has ai/jobs/26262136-ai-training-for-creative-writers-english-japanese-hindi?utm_campaign=google_jobs_apply&utm_source=google_jobs_apply&utm_medium=organic” title=”” rel=”noopener noreferrer” target=”_blank”>aware jobs for contract writers who have master's or doctoral degrees in Hindi and Japanese. Outlier has job postings that list requirements such as academic degrees in mathematics, chemistry, and physics.
“What really makes ai useful to its users is the human layer of data, and that really needs to be done by smart, skilled humans and by humans with a particular degree of expertise and a creative bent,” said Willow Primack, vice president. of data operations in Scale ai. “As a result, we have focused on contractors, particularly in North America.”
Alynza Fenske, a self-published fiction writer, had never interacted with an ai chatbot before hearing a lot from other writers who considered ai a threat. So when she came across a video on TikTok about data annotation technology, part of her motivation was simply to learn everything she could about ai and see for herself if the fears surrounding ai were justified.
“Now that I've been working with it, it's giving me a completely different view,” said Fenske, 28, who lives in Oakley, Wisconsin. “It's comforting to know there are human beings behind this.” Since February, her goal is to put in 15 hours of data annotation work each week to support herself while she pursues a career as a writer.
That Agboh, 28, a master's student in computer science at the University of Arkansas, was tasked with coding projects, for which he paid between $40 and $45 an hour. He would ask the chatbot to design a motion sensor program that helps gym-goers count their reps and then evaluate the computer codes written by the ai. In another case, he would load a set of data on grocery items into the program and ask the chatbot to design a monthly budget. He sometimes even evaluated the codes of other annotators, which experts say are used to ensure data quality.
She won $2,500. But her account was permanently suspended by the platform for violating its code of conduct. She didn't receive an explanation, but she suspected it was because she worked while she was in Nigeria, since the site wanted workers based only in certain countries.
That's the fundamental challenge of online gig work: it can disappear at any time. With no one available to help, frustrated contractors took to social media, sharing their experiences on Reddit and TikTok. Jackie Mitchell, 26, gained a large following on TikTok due to her content about side hustles, including data annotation work.
“I understand the appeal,” he said, calling side jobs an “unfortunate necessity” in this economy and “a hallmark of my generation and the generation above me.”
Public records show that Surge ai owns Data Annotation tech. Neither the company nor its CEO, Edwin Chen, responded to requests for comment.
It is common for companies to hire contractors through subsidiaries. They do this to protect the identity of their clients and helps them avoid the bad press associated with the working conditions of their low-paid contract workers, said James Muldoon, a management professor at the University of Essex whose research focuses on working with data. of ai.
Most of today's data workers depend on salaries from their work. Milagros Miceli, a sociologist and computer scientist who researches working conditions in data work, said that while “many people do this for fun, because of the gamification it entails,” a lot of the work is still “done by the workers.” “They really need the money and use it as their main income.”
Researchers are also concerned about the lack of security standards in data labeling. Workers are sometimes asked to address sensitive issues, such as whether certain events or acts should be considered genocide or what gender should appear in an ai-generated image of a football team, but they are not trained in how to make that assessment.
“Basically, it's not a good idea to outsource or crowdsource on security and ethics issues,” Professor Muldoon said. “You need to be guided by principles and values, and by what your company really decides is the right thing to do on a particular issue.”
