Subscribe to Our Newsletter

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn't arrive within 3 minutes, check your spam folder.

Ok, Thanks

Writer announced enterprise-ready features are coming to its chatbots

Writer has launched significant upgrades to its enterprise chat apps, including built-in RAG technology, explainable AI features, dedicated task modes, and enhanced customization options. The features aim to improve accuracy, transparency, and user experience for businesses building AI applications.

Ellie Ramirez-Camara profile image
by Ellie Ramirez-Camara
Writer announced enterprise-ready features are coming to its chatbots
Credit: Writer

Writer recently announced it would enhance its pre-built chatbot Ask Writer with a suite of features including built-in RAG capable of analyzing up to 10 million words, explainable AI features, dedicated modes, voice rewrites, and custom instructions. The features are also available for custom chatbots built using Writer's AI Studio. The new enhancements aim to make AI more powerful, transparent, and user-friendly for businesses.

Built-in graph-based retrieval-augmented generation (RAG) allows users to analyze an input of up to 10 million words, or approximately 20,000 pages. RAG is more efficient and accurate than an LLM with a large context window, for research shows that on the latter approach, performance degrades as the complexity of the task increases. With RAG built into the chatbots, data is first processed into a knowledge graph preserving the semantic connections between data points. Then, when queries are analyzed, Writer can retrieve the most semantically relevant data points for the query before passing it on to an LLM so it can generate an accurate response.

To further increase the trustworthiness of its chatbot responses, Writer is introducing explainability features that enable chatbots to decompose vague, complex, or broad queries into smaller, more manageable steps for which it can provide answers supported by specific excerpts from the user-provided data. The explainability features deliver transparency, but also insights into the efficacy of their prompts, enabling users to refine their prompts to obtain increasingly better responses.

Moreover, to depart from the one-size-fits-all approach followed by off-the-shelf LLMs featuring a single interface, the Writer update introduces dedicated modes, which are interfaces including a General mode for general-purpose tasks such as brainstorming, text generation, or general knowledge queries; a Document mode for research and drafting tasks; and a Knowledge Graph mode that connects chat apps with data sources to obtain responses based on large corpora of proprietary data. Additional features include rewrites based on voice profiles, custom instructions for consistent formatting, and dictation capabilities.

The new features are aimed at unlocking key enterprise use cases, such as analyzing lengthy documents, completing RFPs, and conducting deep dives on sales prospects. As Writer continues to build up its full-stack AI platform, the company enables its growing user base, which includes organizations like Vanguard, Intuit, and Accenture to build and deploy impactful AI applications.

Ellie Ramirez-Camara profile image
by Ellie Ramirez-Camara
Updated

Data Phoenix Digest

Subscribe to the weekly digest with a summary of the top research papers, articles, news, and our community events, to keep track of trends and grow in the Data & AI world!

Success! Now Check Your Email

To complete Subscribe, click the confirmation link in your inbox. If it doesn’t arrive within 3 minutes, check your spam folder.

Ok, Thanks

Read More