Beyond public access in LLM pre-training data

Summary

Researchers tested whether OpenAI models recognise non-public, pay-walled books they were not licensed to train on, using the DE-COP membership inference attack on 34 copyrighted O'Reilly Media books (13,962 paragraphs). GPT-4o scored an AUROC of 0.82 (95% bootstrapped CI: 0.60 to 0.96), while GPT-3.5 Turbo and GPT-4o Mini scored near chance at 0.54 and 0.56. The authors caution that results are preliminary given the small book sample and differences across models.

Related items

• InfoHumans are still doing the work behind some AI agentsSame vendor · CNBC Technology
• MediumAppraisal by AI: Model-driven victim prioritization in the VOLTA infostealerSame vendor · Cisco Talos Blog
• MediumA Shared ChatGPT Link Could Run Attacker Code in Your Session and Read Private ChatsSame vendor · The Hacker News
• InfoSam Altman says it’s better for RAM to be in a data center than on your deskSame vendor · The Verge (AI)
• InfoThe biggest names in AI are backing a new way to opt out of scrapingSame vendor · The Verge (AI)