// HACKER NEWS — CYBERSECURITY
People Training OpenAI's AI Fired for Using AI to Train the AI
OpenAI has an army of contractors who read real ChatGPT users’ prompts and other data to help improve the chatbot’s responses. The idea is that the contractors provide an, obviously, human touch to OpenAI’s models. Well, not all of the contractors are doing that. 404 Media has found multiple contractors hired to improve OpenAI’s models have been fired for using AI to train the AI. That’s not great for the models themselves, but there is also obviously a great irony in AI training companies working for OpenAI firing people for using AI when OpenAI’s whole thing is to make people use AI at work.
Some AI models already exhibit signs of “model collapse,” which is where AI models further trained on AI-generated text can become worse and worse. In this case, some of the people hired to partially stop that happening are themselves using AI-generated responses to train OpenAI’s models.
One contractor said they see people using AI “all the time and people are let go for it all the time, it’s pretty much the one thing that will get you kicked off ASAP.” The person said, “in a group of thousands there are tons that have been caught.”
Last week, 404 Media revealed Project Lily, in which OpenAI has hundreds of contractors reading real ChatGPT users’ prompts and conversations which can include personal information. Those contractors then rate and critique the responses ChatGPT generated, including making sure that the responses are not too sycophantic or anthropomorphize ChatGPT. That reporting was based on internal documents and conversations with a person who works on the prompts.
404 Media has now obtained other internal documents and spoken to three contractors doing work for OpenAI across various projects. Those projects can include more than ten thousand contractors, according to one of the internal documents.
One of those documents says contractors must not use AI themselves for their work.
“Do not use AI detection tools, or AI yourself,” one document describing the work of contractors who are hired to review the work of other contractors, including catching them for using AI, says. “Do not use GPTZero or any other AI detection tool. They are not reliable. Reviewers may not use AI either, including Grammarly and AI translation, to review, write feedback, or write comments.”
The document continues, “Do not tell evaluators why you suspect AI. It is easier for them to hide if they know what you look for. Judge the overall pattern, not one clue.”
All three of the contractors said reviewers are told not to use AI in their work. Two of the sources said people have been fired or offboarded for using AI. 404 Media granted the contractors anonymity as they weren’t permitted to speak to the press.
The contractors who review other contractors’ work are told to be on the look out for tell-tale signs of AI use. That can include repetitive words, AI-style punctuation — which might include over zealous use of the em dash — and contractors finishing their work very quickly.