• Apple's AI research team has clarified that they did not use YouTube data to train their language models. The company stated that they only used publicly available datasets and did not access any private user data from YouTube or other platforms.
• The clarification came after a researcher claimed that Apple's language models were trained on YouTube data, which raised concerns about potential privacy violations. Apple emphasized that they are committed to protecting user privacy and only use data sources that are publicly available.
• The article notes that Apple's AI research team has been transparent about their work and has published several papers detailing their approach to developing language models. The company's commitment to privacy is a key part of its overall strategy, and it continues to work to ensure that its AI systems are developed in an ethical and responsible manner.