TapTechNews July 18th news, Apple released a statement through the tech media 9to5Mac, regarding the controversy of its OpenELM open-source AI model using YouTube resources for training, stating that OpenELM is not used in any other AI or machine learning projects (including AppleIntelligence).
The non-profit news studio ProofNews released an investigation report, stating that when Apple trained the AI model OpenELM, it used a data set named YouTubeSubtitles with a size of 5.7 GB (489 million words).
This data set was created by EleutherAI and was first released in 2020, involving the subtitle content of 173,536 YouTube videos from more than 48,000 channels, including the subtitle content of more than 12,000 videos that have been deleted from the platform.
In the latest statement, Apple said that the purpose of the OpenELM model is to contribute to the research community and promote the development of open-source large language models.
Apple's researchers once described OpenELM as the most advanced open language model.
Apple emphasizes that OpenELM is only used for research purposes, and not for any commercial AppleIntelligence functions. The model is released in an open source manner, and developers can freely obtain it.
Apple also said that there is currently no plan to build a new version of the OpenELM model.