OPENAI IS SUED FOR USING People’s DATA TO TRAIN AI

A recent complaint was filed against ChatGPT inventor OpenAI, who claims that the buzzy Silicon Valley firm’s AI training practices infringed the privacy and intellectual property rights of everyone who posted anything online.

OpenAI used an enormous quantity of data pulled from all corners of the web to train its sophisticated AI language models. Although OpenAI isn’t sure what its algorithms are trained on, the datasets include anything from articles on Wikipedia and classic literature to postings on social media posts and extremely specific obscenity — and OpenAI didn’t ask for permission for any of it. The collective suit, filed in California, claims that failing to follow basic procurement procedures, such as obtaining the consent of people who created the material in the first place, amounts to data theft.

However, whether the argument stands up in court remains to be seen. The internet’s infrastructure is complicated, and what is commonly referred to as the free and open web is frequently neither; websites have their own user terms and agreements, and even if people have done the work to populate those sites with content, it practically belongs to the platforms — and not, unfortunately, to the users.