Recently, The New York Times filed a lawsuit against ChatGPT. The American media accused OpenAI of misusing its articles without consent. But this matter takes an unexpected turn. In fact, Sam Altman's company communicated very specific arguments.
The whole world was surprised by the complaints filed by The New York Times on December 27th. According to these media outlets, OpenAI used its work to ” develop and market their artificial intelligence products without permission from the Times “. But OpenAI rejected the allegations of the New York Times. The company published an official letter with surprising explanations. According to the AI giantThe Times created the prompts from scratchto retrieve information about your items.
A prompt manipulation that blames OpenAI
“It is interesting to note that the rejections caused by The New York Times appear to be due to old articles distributed on numerous third-party websites. It appears that They intentionally manipulated the promptswhich often include long excerpts from articlesto restore our model. Even if they use such prompts, Our models generally do not behave the way The New York Times suggests“suggesting that they either asked the model to shuffle or chose their examples from multiple trials” from the OpenAI letter.
This statement is an unexpected turn of events for those involved. However, OpenAI claimed this ChatGPT tends to “regurgitate” information.. In rare cases, AI copies newspaper articles “word for word” . But Sam Altman's company has announced that it will fix this error as quickly as possible. OpenAI specialists have already done it The “Browse” function has been removedby ChatGPT. In your opinion, this option is the source of the “burping” data.
According to OpenAI, ChatGPT needs the data to train
Even after the New York Times accusation, OpenAI maintained its position. According to the company, their AI always has to learn to solve new problems. To do this, she must have access to “the vast database of human knowledge.”
However, OpenAI is committed to respecting copyright law . And this approach is already progress for a high-tech company. Interested parties can then request an option to opt out of ChatGPT training. You have to fill out a specific form.
Sam Altman's company isn't closing the doors to collaborations with The New York Times
Despite these trials OpenAI remains open to all forms of partnerships with New York media. They intend to continue negotiations to reach a conclusion. Next, they want to start a long-term collaboration with the Times.
” We hopea constructive partnership with the New York Timesand respect its long history,” OpenAI said in its press release.
This combination could lead to extraordinary results. We're excited to see what happens next. For now, stakeholders appear to be holding their ground.