We want to hear from you! Take our quick AI poll and share your insights on the current state of AI, how you’re implementing it, and what you expect to see in the future. Find out more
Writer, a leading enterprise AI platform, has released a suite of powerful enhancements to its AI chat applications, announced today at VB Transform. The extensive enhancements, which include advanced graph-based generation with advanced retrieval (RAG) and new AI transparency tools, will be live in the Writer ecosystem starting tomorrow.
Both users of the out-of-the-box Writer “Ask Writer” app and developers using the AI Studio platform to build custom solutions will have immediate access to these new features. This widespread adoption marks a significant leap forward in making complex AI technology more accessible and efficient for businesses of all sizes.
At the heart of the upgrade is a dramatic expansion of data processing capabilities. Updated chat apps can now ingest and analyze up to 10 million words of company-specific information, allowing organizations to leverage their own data on an unprecedented scale when interacting with AI systems.
Unleashing the power of 10 million words: How Writer’s RAG technology is transforming enterprise data analytics
“We know that enterprises need to analyze very long files, work with long scientific articles or documentation. This is a huge use case for them,” said Deanna Dong, head of product marketing at Writer, in an interview with VentureBeat. “We use RAG to do actual knowledge mining. Instead of giving to [large language model] LLM the whole library, we’re actually going to do some research, pull out all the right notes and just give the LLM the right resource notes.’
Countdown to VB Transform 2024
Join enterprise leaders in San Francisco July 9-11 for our flagship AI event. Connect with peers, explore the opportunities and challenges of Generative AI, and learn how to integrate AI applications into your industry. Register now
A key innovation is Writer’s graph-based approach to RAG, which maps the semantic relationships between data points instead of relying on simpler vector extraction. According to Dong, this allows for more intelligent and targeted information retrieval:
“We break the data down into smaller data points and actually map the semantic relationship between those data points,” she said. “So a security shard is tied to this architecture tidbit, and it’s actually a more relational way that we map the data.”
A Peek into the Mind of AI: Writer’s ‘Thought Process’ Feature Brings Unprecedented Transparency to AI Decision Making
This graph-based RAG system is at the heart of 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 is taking, including how it breaks down queries into subqueries and which specific data sources it references.
“We show you the steps I take,” Dong explained. “We take something like maybe a potentially broad question or not a super specific question that people are asking, we actually break it down into sub-questions that the AI assumes you’re asking.”
May Habib, CEO of Writer, highlighted the importance of these achievements in a recent interview with VentureBeat. “RAG is not easy,” she said. “If you talk to CIOs, VPs of AI, like anyone who’s tried to build it themselves and care about accuracy, it’s not easy. In terms of benchmarking, a recent benchmark of eight different RAG approaches, including the Writer Knowledge Graph, puts us first in accuracy.”
Personalized AI experiences: Writer’s new “modes” streamline the adoption of AI in the enterprise
The upgrades also introduce special “modes”—specialized interfaces for different types of tasks such as general knowledge queries, document analysis, and working with knowledge graphs. This aims to simplify the user experience and improve output quality by providing more personalized prompts and workflows.
“We observe customers struggling to use a universal chat interface to complete any task,” Dong explained. “They might not guess correctly and get the right results, they forget to say, ‘Hey, I’m actually looking at this file,’ or ‘I actually need to use our internal data for this answer.'” And so they got confused.
Industry analysts see Writer’s innovations as potentially game-changing for enterprise AI adoption. The combination of massive data ingestion, advanced RAG, and explainable AI addresses several key hurdles that have made many firms hesitant to widely deploy LLM-based tools.
The new features will be automatically available in Writer’s pre-built Ask Writer chat app, as well as in all custom chat apps built on the Writer platform. This broad availability can accelerate the integration of AI into various enterprise functions.
“All of these features — the modes, the thought process, you know, the ability to have a built-in RAG — will make this whole suite of fairly complex technologies very usable for the end user,” Dong said. “The CIO is going to be kind of amazed by the built-in RAG, but the end user—you know, the operations team, the HR team—they don’t have to understand any of it. What they’re really going to get is accuracy, transparency, usability.”
As enterprises grapple with how to responsibly and effectively use AI, 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 truly bridge the gap between the enormous potential of AI and the practical realities of enterprise deployment.
