Anthropic Embeds Claude AI Inside Microsoft PowerPoint for Slide Creation
Claude Comes to Office Productivity
On February 23, Anthropic launched Claude in PowerPoint via Microsoft AppSource. The Microsoft 365 add-in embeds Claude AI directly inside PowerPoint, enabling users to generate new slides, edit existing ones, and build complete presentation structures from natural-language prompts — all while respecting the deck's existing slide master, layouts, fonts, and color schemes.
Key Features
- Slide generation: Create competitive comparisons, product overviews, and more from prompts
- Slide editing: Modify specific slides or text elements using natural language
- Brand consistency: Automatically follows existing slide masters and design rules
- Model choice: Switch between Sonnet 4.5 (lighter edits) and Opus 4.6 (complex restructuring)
Limitations and Security
The add-in remains in beta: advanced PowerPoint features lack full support, chat history does not persist between sessions, and Team/Enterprise data retention settings are not yet inherited. Anthropic warns users to be cautious of prompt injection risks when using untrusted files or templates.
Enterprise Expansion Strategy
Claude for Excel is also available in beta for Max, Team, and Enterprise users. The PowerPoint integration is part of Anthropic's broader strategy to embed Claude across productivity tools including Slack, Figma, and Asana.
Source: gHacks
Related Articles
Hacker News treated Anthropic’s Claude Code write-up as a rare admission that product defaults and prompt-layer tweaks can make a model feel worse even when the API layer stays unchanged. By crawl time on April 24, 2026, the thread had 727 points and 543 comments.
Hacker News focused on the ambiguity around Claude CLI reuse: even if OpenClaw now treats the path as allowed, developers still want a clearer boundary between subscription, CLI, and API usage.
Anthropic put hard numbers behind Claude’s election safeguards. Opus 4.7 and Sonnet 4.6 responded appropriately 100% and 99.8% of the time in a 600-prompt election-policy test, and triggered web search 92% and 95% of the time on U.S. midterm-related queries.
Comments (0)
No comments yet. Be the first to comment!