Retrieval-Augmented Generation

Retrieval-augmented generation (RAG) is a technique that enhances large language models by connecting them to external knowledge sources, allowing them to retrieve relevant information before generating responses. This approach significantly reduces hallucinations and improves accuracy, making it critical for enterprise applications where factual correctness is non-negotiable. For AI governance, RAG creates important compliance considerations around data sourcing, retrieval transparency, and the ability to audit which knowledge bases informed specific model outputs.

1 item

ResearchGlobal2026-09-02

215,000 Fake AI Buying Guides Are Poisoning RAG Procurement Pipelines

Trellner Research found that three coordinated websites published over 215,000 machine-generated software recommendation pages designed to be retrieved by AI systems rather than read by humans. Testing across 380 software categories showed that 59.8% of citations returned by Perplexity pointed to domains outside the top 100,000 most-visited websites. The findings expose a direct integrity risk in any enterprise workflow that uses AI-grounded search to inform vendor selection or procurement research.

Perplexity RAG integrity AI sourcing transparency grounding manipulation AI procurement risk retrieval-augmented generation vendor due diligence