Claude Adds Live Dashboards and AI Video Generation Tools

What happened

Anthropic has added two new beta capabilities to Claude: Dashboards, which turns connected data sources such as BigQuery and Snowflake into live, interactive dashboards from a text prompt, and Motion, which generates animated explainer videos from text and image inputs. Alongside the launch, Anthropic has also widened access to its existing Docs, Slides and Design tools, which are now available across all Claude plans, including free accounts.

Dashboards is aimed at turning raw, connected datasets into visual, queryable views without a user needing to build charts or write queries manually. Motion extends Claude's output beyond static documents and slides into short animated video content, generated from written prompts and supplied imagery.

Why it matters

The update signals Anthropic's continued push to move Claude from a conversational assistant into a broader productivity and content-generation layer that sits across data analysis, document creation and now video output. By connecting directly to enterprise data warehouses, Dashboards pushes generative AI further into business intelligence territory traditionally occupied by dedicated BI tools, lowering the technical bar for turning live data into something decision-makers can actually look at.

Making Docs, Slides and Design free-tier features also matters for adoption: it removes a paywall that previously limited experimentation with Claude's productivity suite to paying users, which could accelerate how quickly these tools get tested inside organisations and teams that haven't yet committed budget to AI platforms.

The Renascence take

The more interesting story here isn't the video generator — it's what happens when "ask a question, get a dashboard" becomes normal. That shift changes who gets to interrogate data, and when.

Most coverage will focus on the novelty of AI-generated explainer videos, but the sharper signal is Dashboards: live BI generated from a prompt collapses the distance between a question and an answer, which is exactly the kind of friction behavioral economics tells us shapes whether people bother to look at data at all. Organisations that plug this into real BigQuery or Snowflake environments should treat it as a governance exercise as much as a productivity one — the risk isn't that the dashboard is wrong, it's that it's instantly trusted because it looks finished. Teams should pilot this on low-stakes, non-customer-facing data first, and be deliberate about who gets prompt-level access to live data before this becomes shadow BI across the business.