The Register: OpenAI wants to bend copyright rules. Study suggests it isn’t waiting for permission

Apr 3, 2025

—

Source URL: https://www.theregister.com/2025/04/03/openai_copyright_bypass/
Source: The Register
Title: OpenAI wants to bend copyright rules. Study suggests it isn’t waiting for permission

Feedly Summary: GPT-4o likely trained on O’Reilly books without permission, figures appear to show
Tech textbook tycoon Tim O’Reilly claims OpenAI mined his publishing house’s copyright-protected tomes for training data and fed it all into its top-tier GPT-4o model without permission.…

AI Summary and Description: Yes

Summary: The text discusses allegations made by Tim O’Reilly regarding OpenAI’s use of O’Reilly’s copyrighted books for training the GPT-4o model without authorization. This issue relates to privacy and copyright laws, presenting implications for compliance as it highlights the need for clear guidelines regarding data sourcing in AI model development.

Detailed Description: The claim made by Tim O’Reilly sheds light on the complexities and legal concerns surrounding AI training datasets. The significant points include:

– **Copyright Concerns**: O’Reilly’s assertion that OpenAI utilized copyrighted material without permission raises questions about intellectual property rights in the realm of AI training.
– **Training Data Sources**: The ethics involved in sourcing data for training AI models come under scrutiny, leading to potential legal consequences for organizations that do not obtain proper licensing.
– **Compliance Implications**: This case underscores the necessity for organizations involved in AI development to establish robust compliance frameworks to ensure they respect copyright laws while leveraging data.
– **Industry Feedback**: The incident may tighten industry standards, pushing for more transparent disclosures of how training data is gathered and used, fostering an environment of better governance and accountability.

Overall, this issue is a critical reminder for security and compliance professionals to stay informed about copyright laws and the intricacies of using third-party content in AI training processes while emphasizing the need for transparent practices in data sourcing.

-4o -4o model 2 2025 3 4 5 a account accountability Act AGI AI AI development ai model AI models and app art as authorization AWS by bypass C CERN CI CleaR co compliance compliance framework compliance frameworks compliance implications compliance professionals concerns content copyright copyright concerns Copyright Law copyright laws core critical D data data sources data sourcing dataset datasets de development disclosure e end environment Ethics feedback for framework frameworks g GIS Go governance governance and accountability GPT GPT-4o guidelines H high Highlight http HTTPS implications in incident industry industry standards Intel Intellectual Property Intellectual Property Rights Iron k l law led Legal legal concerns legal consequences Li licensing lm mission Mode model model development models N no o of on open openai OPM organization organizations out over party point potential pre privacy process processes professionals Property Rights publishing Py Q question R rag RCE real red right Ro RoT s sec security security and compliance sequence Sig source sourcing SSE standards study T tech text the third third-party to TP training training data training datasets training processes transparent UI under US use uth V WAN Wi x