Writer Unveils Mind-Blowing AI Update: RAG on Steroids, 10 Million Word Capacity, and AI’s ‘Thinking Process’ Revealed

Writer, a leading enterprise AI platform, has launched a series of powerful enhancements to its AI chat applications, announced today at VB Transform. These sweeping improvements, which include advanced graph-based retrieval (RAG) augmented generation and new tools for AI transparency, will be rolling out across the Writer ecosystem starting tomorrow.

Users of Writer’s standard Ask Writer app and developers leveraging the AI ​​Studio platform to create custom solutions will have immediate access to these new capabilities. This broad rollout marks a significant step forward in making sophisticated AI technology more accessible and effective for businesses of all sizes.

At the heart of this upgrade is a dramatic expansion of data processing capabilities. The redesigned chat applications can now digest and analyze up to 10 million words of business-specific information, enabling organizations to leverage their proprietary data at an unprecedented scale when interacting with AI systems.

Unlocking the Power of 10 Million Words: How Writer’s RAG Technology is Transforming Enterprise Data Analysis

“We know that companies have to parse very long files, work with long research papers or documentation. That’s a huge use case for them,” Deanna Dong, product marketing manager at Writer, said in an interview with VentureBeat. “We use RAG to do knowledge retrieval. Instead of giving the LLM (large language model) the whole library, we’re actually going to do a search, pull out all the good notes, and just give the LLM the good resource notes.”

Writer’s key innovation is its graph-based approach to RAG, which maps semantic relationships between data points rather than relying on simpler vector retrieval. According to Dong, this enables more intelligent and targeted information retrieval:

“We break down the data into smaller data points and we map the semantic relationship between those data points,” she said. “So a piece about security is linked to this piece about architecture, and it’s actually a more relational way of mapping the data.”

A Look Inside the AI ​​Mind: Writer’s ‘Thinking Process’ Feature Brings Unprecedented Transparency to AI Decision-Making

This graph-based RAG system leverages a new “thought process” feature that provides unprecedented transparency into how the AI ​​arrives at its answers. The system shows users the steps the AI ​​takes, including how it breaks down queries into sub-questions and the specific data sources it references.

“We show you the steps it takes,” Dong explained. “We take a sort of potentially broad or not-so-specific question that people ask, and we break it down into sub-questions that the AI ​​assumes you’re asking.”

Writer CEO May Habib emphasized the importance of these advances in a recent interview with VentureBeat. “RAG is not easy,” she said. “If you talk to CIOs, VPs of AI, just about anyone who’s tried to build it themselves and cares about accuracy, it’s not easy. In terms of benchmarking, a recent head-to-head test of eight different RAG approaches, including Writer Knowledge Graph, we came out on top in terms of accuracy.”

Tailored AI Experiences: Author’s New ‘Modes’ Streamline Enterprise AI Adoption

The upgrades also introduce dedicated “modes,” specialized interfaces for different types of tasks such as general knowledge queries, document analysis, and working with knowledge graphs. The goal is to simplify the user experience and improve the quality of results by providing more personalized prompts and workflows.

“We see that customers struggle to use a universal chat interface to do every task,” Dong says. “They don’t always respond accurately and get the right results. They forget to say, ‘Hey, I’m looking at this file,’ or ‘I need to use our internal data for this response.’ So they get confused.”

Industry analysts see Writer’s innovations as a potential game changer for enterprise AI adoption. The combination of massive data ingestion, sophisticated RAG, and explainable AI overcomes several key hurdles that have made many companies hesitant to deploy LLM-based tools at scale.

The new features will be automatically available in Writer’s pre-built “Ask Writer” chat app, as well as any custom chat apps built on the Writer platform. This broad availability could accelerate the integration of AI into various business functions.

“All of these features – the modes, the thinking process, you know, the ability to integrate the RAG – are going to make this very sophisticated set of technologies very usable for the end user,” Dong said. “The CIO is going to be pretty impressed with the integrated RAG, but the end user – you know, an operations team, an HR team – doesn’t need to understand any of this. What they’re really going to get is accuracy, transparency, ease of use.”

As businesses look to leverage AI responsibly and effectively, Writer’s latest innovations offer a compelling vision for more transparent, accurate, and user-friendly LLM applications. The coming months will reveal whether this approach can indeed bridge the gap between AI’s immense potential and the practical realities of enterprise deployment.