Claude is no longer a chatbot: Anthropic adds data dashboards and video to compete for the office

By Unite.AI · October 9, 2026.

Anthropic introduced two new tools inside Claude on October 8 —Dashboards and Motion— and, the same day, took Claude Docs, Slides and Design out of beta across all plans, including the free tier, as reported by Unite.AI based on the company's announcement. The company also set December 14, 2026 as the shutdown date for claude.ai/design, the standalone site that had been running alongside the chat.

Dashboards, available in beta on paid plans, connects Claude to a company's data warehouse —Amazon Redshift, BigQuery, ClickHouse, Databricks or Snowflake— and to connectors such as Salesforce. The user asks in natural language (for example, comparing one week's signups with the previous month's) and Claude builds the dashboard, keeps it updated and shows the exact query behind each chart, along with the date it was last refreshed. When the analysis calls for more depth, the dashboard can be sent to specialized tools —Amplitude, Grafana, Hex, Mixpanel, Omni, Perplexity, PostHog or Sigma— with Looker, monday.com and Tableau announced as upcoming integrations.

Motion, in beta for Team and Enterprise plans, turns a prompt into an animation that plays as a short video —among Anthropic's examples, a quarterly report turned into a 30-second explainer for an all-hands, a chart set in motion for a board presentation— which can then be edited in an editor or by asking Claude, and downloaded as an MP4 file. Anthropic highlights a technical detail with editorial weight: Motion does not use a video generation model, but instead writes code that animates text, charts, shapes and the user's own images, so there is no generated footage and no synthetic people. Finished animations can be opened in Adobe, Descript, HeyGen, Higgsfield, invideo, Luma AI or Runway, with Canva and Captions coming next.

Docs, Slides and Design —which arrived in beta on September 16, when Anthropic merged Claude Cowork into the main chat— have been used, according to the company's own figures, to create more than 45 million documents, presentations and designs in Claude. Coming out of beta adds customer-managed key encryption (CMEK, the standard that lets an organization control its own encryption keys instead of handing them over entirely to the provider), template control for administrators, collaborative editing alongside Claude, export to PowerPoint and PDF with formatting preserved, export of presentations to Google Slides as editable files, and editing from mobile.

The company has been careful with the switches: on Enterprise accounts, Dashboards and Motion arrive off by default and must be enabled from the organization's settings, while Docs, Slides and Design will turn on automatically on October 15 unless the administrator enables them sooner. Organizations with zero data retention cannot enable artifacts or migrate their projects, so Anthropic is asking them to download their work from the standalone site before it closes in December; everyone else can migrate their design systems from the Artifacts page, although the chats and comments tied to the original Design will be deleted after the shutdown, with no exact timeframe specified.

Our reading: this is not a round of scattered features, it is Anthropic building the entire office inside a chat window. A BI dashboard (business intelligence, the analysis of corporate data for decision-making) that reads straight from Snowflake or Salesforce, a presentation that gets animated without leaving Claude and a document that exports to PowerPoint without losing its formatting are, each on its own, reasonable conveniences. Together, they are an attempt to make the analyst, the deck designer and the basic video editor stop needing Tableau, Canva or Adobe as a starting point, and start —and finish— in Claude.

This fits with something we have already been pointing out about the war among the AI giants: the decisive ground has shifted from the smartest model to who controls the "plumbing" of daily work —distribution, integrations, agent standards. Google and Microsoft are ahead there because they already have Workspace and Office inside a huge number of accounts; Anthropic, which has no office operating system of its own, is building one on top of the chat, weaving connectors out to the tools that do dominate each niche (Grafana, Runway) instead of taking them on head-on. It is a smart challenger strategy: not to replace the enterprise software ecosystem, but to become the conversational layer that organizes it.

On the jobs front, which we follow sector by sector, this directly reduces part of the routine work of the junior data analyst, the presentation designer and the explainer video editor —the kind of administrative, repetitive task that, as we have argued, is the most exposed in the short term. The decision that Motion should not use a video generation model, but auditable, editable code, is a sign of responsible design worth acknowledging: it avoids the synthetic-people problem and keeps the animation under human control, in line with governance based on evidence and not on panic. At the same time, plugging a conversational assistant straight into a company's live data warehouse —with permissions, its own encryption and carefully staggered default activation— is exactly the kind of integration that demands serious governance from whoever administers it, not just from the provider.

In the long run, if Claude delivers on its promise, the cost of producing a well-grounded analysis, a clear presentation or an explainer video drops noticeably for anyone, not just for those who know how to handle Tableau or Adobe Premiere. That is, on a small scale, the abundance we talk about when we think about AI's horizon: less time spent producing the supporting material for the work, more time spent deciding what is worth analyzing or telling.