Traditional Knowledge Management has long been plagued by documentation debt, the critical lag between case closure and knowledge creation that leads to fragmented insights and redundant content. Informatica solved this challenge by transitioning from manual repositories to a dynamic, agentic framework. They have reached a defining 80/20 milestone, where AI agents now generate 80% of knowledge articles with 75% baseline accuracy that also drives 60% of case deflection. This shift has also redefined the role of their human experts, moving them from primary authors to strategic reviewers. By freeing those engineers from the burden of manual drafting, they can now focus on solving complex edge cases and enriching the knowledge base with high-level expertise only when needed.

Historically, knowledge management programs and technology for support organizations had one goal: to automate the common, repetitive problems, so support engineers could focus more time on new and unique issues. The general assumption used the Pareto Principle that 80% of issues had occurred before, while 20% of issues were new. The role of knowledge management and automated search technology was to create and find content for those 80% of issues, and let agents run the play for the remaining 20%. But for many B2B enterprise technology companies, this strategy has failed to achieve sufficient success as product support became more complex.

How can confidential information like PII, financial data, and intellectual property be protected in an artificial intelligence world?

Aaron Fulkerson, founder of Mindtouch, former ServiceNOW GM, and current CEO of Opaque Systems, discusses privacy-preserving solutions, even in highly regulated industries like financial services and healthcare. 

SPAR Solutions founder Swami Ganapathy also discusses some of the challenges that his clients are experiencing alongside powerful solutions that any organization can adopt.

Now that we’ve completed the 101 and 201 classes for Gen AI, let’s go beyond the hype and talk about tactical, real-world uses of implementing Generative AI within knowledge management and support, including concepts like:

1. How to automatically summarize information from multiple documents and present to a user in self-service rather than relying on an agent to do this manually.

2. How to make your existing content more personalized so that it is better consumed on self-service, including real-time language translation, making a technical article less technical for a layman to better understand, and more.

3. How Generative AI can turbocharge your KCS program to automatically write knowledge articles based on the information from closed cases.

Our last session, Inside the Minds of Generative AI, was the most popular yet. So we followed that up with this 201-level class, including a deeper dive into LLMs and an overview of Vector Databases, Retrieval Augmented Generation (RAG), and more.

What exactly is generative AI, how does it actually work, is it truly intelligent, why does it hallucinate, and how can businesses use this technology? In this KM Best Practices session, some of the best minds in the space address these and many other questions while exploring the good, the bad, and the ugly use cases of ChatGPT-like technologies. Whether you are an AI enthusiast or a business leader looking to stay ahead of the curve, this session provides valuable insights into one of the most exciting developments in the world of AI.