Anthropic’s Claude Tag is learning your company, one Slack message at a time

Anthropic has launched Claude Tag in a research preview, marking a significant step forward in AI-assisted teamwork within Slack. This innovative feature acts as an 'always-on' AI partner, allowing users to tag @Claude for insights during conversations and for task assignments. Initially available to Claude Enterprise and Claude Team customers, this tool builds on existing functionalities that let users directly message Claude or tag it in channels for immediate assistance. What sets Claude Tag apart is its ability to maintain persistent context and memory, a feature that the previous integrations struggled with. According to Anthropic, as Claude interacts in a Slack channel, it continuously learns about the ongoing work. It can also collect information from other parts of the organization, provided it has permission to access those channels. With Claude Tag, all members of a Slack channel can engage with a unified Claude identity. This means that anyone can view Claude's previous contributions and seamlessly continue discussions from where the last user left off. System administrators will determine the specific tools and channels Claude can access, ensuring that the AI maintains relevant information tailored to its designated tasks. For instance, a Claude configured for legal tasks will not disrupt discussions in engineering channels. When assigned a task, Claude Tag methodically breaks it down into manageable stages, utilizing its available tools to produce results shared directly within Slack threads. Moreover, it features an ambient mode that allows it to proactively engage in conversations, keeping teams informed, flagging important updates from the organization, and following up on tasks that may have been overlooked. Anthropic's approach aims to create an experience akin to collaborating with a real colleague. This enhanced contextual awareness is becoming an essential focus for enterprise applications, with other companies like Microsoft also emphasizing this aspect through solutions like Graph and Copilot. Platforms such as Snowflake and Databricks are similarly paving the way by integrating organizational knowledge that AI agents can leverage, while Glean is developing an intelligence layer that bridges AI models with enterprise data.