Top news
Investing in the public benefit OfficialAnthropic NewsOct 18Popular Musk Pledges SpaceX Shares to ‘Every Kid in America’ WebBloomberg TechnologyOct 19Popular Websites are closing their doors to AI agents WebThe Verge AIOct 17 This is the new ChatGPT Voice, powered by GPT Live OfficialOpenAI on YouTubeOct 18 How AI giants will reshape global philanthropy WebAxios AIOct 19 Gemini 4 Argon Really Surprised Me WebTwo Minute PapersOct 19 Google Gemini 4 agents just beat Minecraft with no human intervention WebEngadget AIOct 19 Who could it be? 🧐 OfficialGoogle on YouTubeOct 18 Anthropic gives outside AI watchdog “employee-like” access WebArs Technica AIOct 18 What are self-replicating ‘AI worms’, and why is Anthropic worried about them? WebThe Guardian AIOct 18
GlossaryProductsNew to AI

Retrieval augmented generation

Also written: RAG, Agentic RAG, Agentic Retrieval Augmented Generation, Retrieval, RAG-grounded, RAG Engine, Retrieval methods, Information retrieval

Retrieval augmented generation (RAG) enhances AI by finding external information before answering.

In plain terms

Retrieval augmented generation, or RAG, is a way to make AI models smarter by giving them access to extra information. Instead of just relying on what the AI was trained on, RAG lets the AI look up specific details from a set of documents or a database.

For example, imagine you ask an AI about a company's latest financial report. If the AI uses RAG, it will first search for that report and then use the information it finds to give you an accurate answer. This helps the AI provide more current and precise responses than it could otherwise.

The expert version

Retrieval augmented generation (RAG) is a technique that improves the factual accuracy and relevance of large language model (LLM) outputs. It works by retrieving relevant documents or data snippets from an external knowledge base, such as a vector database or document store, and then incorporating this retrieved context into the prompt provided to the LLM. This allows the LLM to ground its responses in specific, up-to-date information, mitigating issues like hallucination and knowledge cutoffs inherent in static training data.

Why it matters now

Recent developments show a strong focus on making AI agents more capable and efficient. Companies are exploring ways to ground AI responses in specific, external data sources, whether for enterprise use cases or to improve local, private AI applications. This includes using specialized tools and retrieving information from databases like SQLite to enhance AI's understanding and accuracy.

Where it shows up 44 items, newest first
How each company uses it
Anthropic Claude
Anthropic is exploring agentic RAG for customer support, enabling AI agents to easily use specialized tools for better performance.
Google Gemini
Google is integrating retrieval methods into its AI, with new embedding models and local RAG recipes using SQLite to improve AI understanding and enable offline processing.
Microsoft Copilot
Microsoft Copilot is now generally available with Fabric IQ, allowing it to reason over enterprise data and provide answers grounded in trusted business context.
Related terms

Large Language ModelsVector DatabasesPrompt EngineeringKnowledge BasesAI AgentsHallucination

Often confused with
Fine-tuning
Fine-tuning retrains a model on new data, altering its internal wei
Embeddings
Embeddings are numerical representations of text used to find simila
Search Engines
Search engines retrieve information, but RAG specifically uses r
Origin

The concept of augmenting generation with retrieved information has roots in earlier information retrieval and question answering systems. The term RAG gained prominence with research papers exploring how to combine retrieval with large language models for improved performance. Source ↗

Try this about 5 minutes

Upload a PDF document to an AI assistant that supports document uploads, like ChatGPT Plus or Claude. Then, ask the AI specific questions about the content of that document. Observe how it answers based on the information you provided.

Try it: Playbook entries that use thisPlaybook →

The Playbook: reusable how-tos, prompts, configs and tips pulled out of the news.

Questions people ask

How does RAG make AI answers better?
RAG helps AI by letting it look up specific facts from your documents before it answers. This makes the answers more accurate and relevant to the information you provided.
Does RAG change the AI model itself?
No, RAG does not change the AI model's core programming. It's like giving the AI a cheat sheet to use during a test, rather than teaching it new material permanently.

First seen on Subvolts Jan 12, 2024. Updated Oct 17, 2026. Written by an AI model from the items above and reviewed against them weekly; report an error.