Andrew Gamino-Cheong is the CTO & Co-Founder of Trustible, and AI Governance startup helping organizations adopt Responsible AI and comply with emerging AI regulations such as the EU AI Act. Prior to starting Trustible, Andrew was a machine learning engineer and tech lead at FiscalNote where he developed and deployed AI systems to analyze the policy making process.
The focus in AI is shifting from LLMs to retrieval-augmented generation, or RAG, which merges data retrieval with language generation.
Article defining AI governance and the various responsibilities of different teams/roles in that process
A lot of use cases of LLMs are limited by data that might be older, and RAG patterns are the most effective way of keeping them up to date without spending millions on fully retraining them. One secret is that a lot of LLM providers would love for users to add RAG pipelines or outright fine-tune their foundational models because it radically shifts a lot of product liability.
RAGs have been used for this application for years before LLMs even appeared on the public’s radar. Overall, practically any application that requires you to have a tightly controlled dataset will favor using an RAG, as they allow for less surprises and much more consistent results across the board.