Slack adds AI-fueled search and summarization features to platform

Slack, a popular platform for enterprise communications, has unveiled new features aimed at making it easier for users to access and summarize information stored within the platform.

Noah Weiss, Slack's chief product officer, highlights the challenge of extracting valuable knowledge from the informal and unstructured data accumulated on the platform over the years. He emphasizes the role of generative AI in unlocking this hidden treasure trove of information.

The introduction of AI-driven features like channel summarization aims to streamline information consumption for users. This functionality enables employees to quickly catch up on missed conversations or grasp the essence of lengthy threads without having to read through them entirely.

Weiss emphasizes the importance of transparency and trust in the design of these features. The AI-generated summaries provide references, allowing users to understand how each part of the summary was created and providing detailed context for further exploration.

In addition to summarization, Slack's AI-powered search capability enables users to ask natural language questions and receive relevant answers derived from Slack content. Each answer includes a quality check feature, allowing users to provide feedback and contribute to the model's improvement.

While Slack did not delve into specifics about the underlying AI model, Weiss mentions that it is a combination of various large language models, fine-tuned for Slack's data and prompt engineering requirements.

These AI features are available as add-ons for Slack's enterprise plans, providing users with enhanced search and summarization capabilities. While specific pricing details were not disclosed, the features are currently accessible in the U.S. and U.K., with plans for additional language support in the future.