The question of whether training AI models like ChatGPT on copyrighted works is legally permissible is a pressing issue in the United States. As of early 2024, there is no clear legal precedent or comprehensive legislative framework specifically addressing this matter. The outcome of this question has significant implications for the AI industry, content creators, and copyright holders.

Several key developments are unfolding that could influence the resolution of this issue:

1. Legislative Initiatives: Congress has shown interest in the intersection of AI and copyright law. Notably, the AI Foundation Model Transparency Act was introduced in December 2023. This act and other legislative proposals could establish guidelines or laws clarifying the use of copyrighted materials in AI training​​.

2. Executive Actions and Recommendations: Following President Biden's Executive Order, the U.S. Copyright Office is expected to issue recommendations by the end of July 2024 on potential executive actions relating to copyright and AI, including the treatment of copyrighted works in AI training. These recommendations could prompt further decisive actions or frameworks​​​​.

3. Court Cases: Several lawsuits against major tech companies, including those involving OpenAI, are currently in progress. These cases have the potential to set legal precedents, depending on how courts interpret and apply copyright law to the use of copyrighted content in training AI models. The outcomes of these lawsuits could emerge in 2024, though court timelines can be variable and subject to delays​​​​.

Given these factors, the question aims to predict whether a major legal decision, either through court rulings or legislative actions, will establish a clear legal precedent regarding the use of copyrighted works for training AI models in the United States by the end of 2024.