EmbeddingGemma 2: an open, lightweight multimodal embedding model
deepmind.google·6d ago
AI Engineering
Hacker News·Aug 9·5 min read
ChatGPT starts blocking direct requests to copy an author's style
TL;DR
OpenAI's ChatGPT now avoids generating text that directly imitates the style of famous authors, instead offering original content that captures their essence. This change may have legal implications amid ongoing copyright lawsuits.
✦ Why It Matters
Review your AI usage policies to ensure compliance with copyright laws and avoid potential legal issues.
Key Takeaways
- Understand the legal implications of AI-generated text.
- Avoid direct style imitation in AI applications.
- Monitor ongoing copyright lawsuits affecting AI models.
- Educate teams on copyright law and AI usage.
Full Summary
OpenAI's ChatGPT has implemented a new policy that prevents it from directly mimicking the writing style of well-known authors, such as Stephen King and J.K. Rowling.
Instead, when prompted, it generates original text that reflects the broader themes and characteristics of those authors' works without copying their distinctive voice. This shift comes as OpenAI faces multiple lawsuits from authors claiming copyright infringement due to the model's ability to produce text similar to their copyrighted materials.
While copyright law protects specific expressions of ideas, the line between inspiration and infringement can be blurred, especially if the AI-generated text is deemed 'substantially similar' to the original. Legal experts note that this is a unique situation, as AI can now imitate personal styles more easily and affordably than ever before.
OpenAI's spokesperson has not commented on this development, which could significantly impact how AI models are trained and used in the future.