Researchers poison their own data when stolen by an AI to ruin results

• Researchers from China and Singapore proposed AURA (Active Utility Reduction via Adulteration) to protect GraphRAG systems
• AURA deliberately poisons proprietary knowledge graphs so stolen data produces hallucinations and wrong answers
• Correct outputs require a secret key; tests showed ~94% effectiveness in degrading stolen KG utility

Researchers from universities in China and Singapore came up with a creative way to prevent the theft of data used in Generative AI.

Among other things, there are two important elements in today’s Large Language Models (LLM): training data, and retrieval-augmented generation (RAG).

Training data teaches an LLM how language works and gives it broad knowledge up to a cutoff point. It doesn’t give the model access to new information, private documents, or fast-changing facts. Once training is done, that knowledge is frozen.