OpenAI Allegedly Uses YouTube Data to Train AI Despite Google’s Policies.

OpenAI has reportedly used YouTube videos to train its speech-to-text AI language model Whisper, despite Google’s rules prohibiting the use of video content for anything other than personal use. The Information reported that OpenAI trained its Whisper using YouTube video content. This information has not been denied by either Google or OpenAI. However, Google has also previously utilised ChatGPT, another AI language model, to help train its Bard AI in the past. The current model of Bard isn’t the product of ChatGPT training, according to Google. Google and OpenAI have declined to comment on the issue. Yann LeCun, Meta Platforms’ AI chief, has emphasised the importance video data plays in training AI. Video data is an important factor in helping train AI to think like humans do. However, text training can also serve as good fodder for some tasks, like writing customer support emails.