People who are paid to train new AI models by supplying them with high-quality conversation and tests are cheating and using chatbots like ChatGPT to do the job instead, multiple whistleblowers have told New 杏吧原创. The seemingly widespread practice risks undermining the future of AI, as it could lead to the 鈥渃ollapse鈥� of more advanced models.
Most AI models operating today were trained on text and data scraped from the internet. But as models have scaled up, requiring yet more training data, AI firms have begun using workers who carry out conversations and tests with AI, in the hope that the resulting high-quality data can improve the power and usefulness of future large language models (LLMs).
These workers are normally employed by third parties, rather than AI companies directly, and are often working without full-time contracts and for low pay. That can incentivise them to take shortcuts like using chatbots to complete tasks faster, according to a worker called Alice*, despite this being against company policies.
Advertisement
鈥淚t鈥檚 very widespread; every company I鈥檝e worked for has had explicit guidelines around it and they clearly do try to catch people out, so I think they do care. But I don鈥檛 think they can stop it,鈥� says Alice.
Alice says she听feels 鈥渘ot in the slightest鈥� guilty about using ChatGPT to complete training tasks, saying it is easy to get away with as long as you instruct chatbots to avoid the usual telltale signs of AI output, . 鈥淚t鈥檚 only the sloppiest of users that get caught,鈥� she says. 鈥淎nyone with a modicum of awareness around AI hallmarks can tell their output not to use them, and at that point what are you going to do?鈥�
鈥淚f these companies want quality data, then they should offer quality contracts,鈥� says Alice. 鈥淚nstead they鈥檙e low-balling struggling people, employing them for the barest possible amount of time and tossing them aside as projects are finished with no warning.鈥�
Another worker, Bob*, worked for a training platform called Outlier. Initially, he was tasked with AI training, which he says he illicitly used AI for, and was then promoted to a leadership role where part of his job was to catch others doing the same thing.
鈥淢anagement vacillated between light tolerance to outright banning,鈥� says Bob. Workers at Outlier would be tracked with a tool called Hubstaff which takes screenshots of their desktop at random intervals to ensure they are really doing tasks as ordered. Bob would look for evidence of AI models in those screenshots.
鈥淧eople would have it [AI models like ChatGPT] open in other tabs, or minimised, so obviously we could see it in the task bar,鈥� says Bob. 鈥淓ven stuff like folders on their desktop with names gave it [AI use] away.鈥�
Outlier, which is owned by Scale AI, did not respond to a request for comment. Scale AI to carry out work for technology giants like Meta and Cisco, neither of which responded to New 杏吧原创鈥榮 request for comment. Bob says he had personally worked on projects for Google, which also did not respond to a request for comment.
Another worker, Carol*, who has worked on several platforms, says that her use of AI began by checking her work for anything that went against the lengthy guidelines for a task, because any contravention could mean expulsion from the project and a loss of earnings.
鈥淚 was terrified of not having an income source, and then after that, it just became easier to run everything through LLMs,鈥� says Carol. 鈥淔or a lot of the projects that I do now, it鈥檚 creating scenarios, so I will use one LLM to help me create the scenario and then I鈥檒l use a different LLM to help me create the files that go along with the scenario. I do feel guilty but like I said, in the beginning it was more about trying to make sure I wasn鈥檛 making any errors.鈥�
鈥淚 do worry that I鈥檓 actually making it [AI] worse. I thought using the models to train themselves negates some of the value,鈥� says Carol.
at the University of Birmingham, UK, says research has shown that AI models 鈥渃ollapse鈥� if they are recursively trained on AI-generated content. When this happens, the abilities of the model drop dramatically and they become less useful. The process is sometimes known as AI cannibalism or AI inbreeding.
鈥淭hat鈥檚 the kind of worst-case scenario. And that鈥檚 probably not what鈥檚 happening in the real world,鈥� says Lee. 鈥淭here鈥檚 still a few humans. And if you have like 10 per cent human data, it mitigates it, it avoids model collapse.鈥�
But Lee says that the kind of cheating these workers are doing isn鈥檛 without repercussions, and will hit performance. 鈥淩ather than it being catastrophic, you鈥檒l see that the AI isn鈥檛 as good at doing human-like tasks. It鈥檚 an issue, because I think the models aren鈥檛 as good as they could be.鈥�
*Names have been changed to protect identities
