Customer feedback is crucial to shaping our product roadmap. We’d like to take this opportunity to share some examples of how your feedback directly resulted in enhancements that we’ve made in the IBM Content Assistant. Prior to going live, we ran a Preview Program, which provided early access for users to experience and provide feedback on all aspects of the IBM Content Assistant: the installation and configuration process, its capabilities and more.

After analyzing the feedback obtained, we developed two new capabilities that are now available in the GA-version of IBM Content Assistant!

1. Generating a summary that is derived from multiple documents: This feature allows users to quickly summarize a a collection of documents in order to grasp the essence of these without having to read through the entire contents. This feature is great for quickly getting a summary of documents in a given folder, or a summary of the documents in a search result set. This capability has resonated particularly with the following use cases:

1. Healthcare agents no longer need to spend time manually reviewing patient files; they can use IBM Content Assistant to rapidly generate a summary related to a patient in order to quickly glean key information, saving significant manual effort.

2. Within the insurance industry, business users can easily generate a summary of claims-related documents for a given customer and quickly familiarize with the key elements of the claim(s). Significant time is saved, since the business users don’t need to spend time individually reviewing each document

2. Customized summaries for each document class: The requirements for a document summary can vary significantly, based upon the document class. For example, a summary of a 100-page contract should be quite different than the summary of a 3-page medical letter. This feature empowers users to tailor the length and substance of the auto-generated summaries to meet their specific requirements on a per-document-class basis. This allows for summaries of different classes to surface different characteristics depending on the use case.

1. Legal documents like policies, for example, can be very dense and lengthy. Using this capability, characteristics like the summary length can be tweaked to accommodate for longer summaries.

2. On the other hand, a medical correspondence letter may be shorter in nature, but may have strict requirements about the nature of the PII that should be included in the summary. The summary generation prompt can be modified to convey the specific nature of what should and should not be included in such a summary, ensuring compliance with strict regulatory and business requirements.

These improvements were a direct response to feedback and suggestions provided during the Preview, demonstrating our commitment to creating products that delights our clients. These enhancements significantly improve the IBM Content Assistant user experience, making it more efficient and effective. We would like to express our sincere gratitude to all our Preview participants. Your active engagement and constructive feedback have been the driving force behind these improvements. We look forward to continuing our dialogue and working with our customers and partners to shape the future of IBM Content Assistant!