Notion AI Releases April 2026: Complete List of Every AI Feature Shipped
Notion AI Releases April 2026
Notion shipped six distinct AI releases in the first two weeks of April 2026. This page catalogs every release, its availability, and its practical impact so you can decide which ones matter for your workflow.
All Notion AI Releases This Month
| Release | Ship Date | What Changed | Availability | Impact Level | |---|---|---|---|---| | Voice input for AI prompts | April 6 | Dictate prompts to Notion AI via a held key on desktop | Plus and above | Medium | | Custom meeting note instructions | Early April | Workspace-level templates for AI meeting summaries | Business and above | Medium | | Cmd+K meeting notes | Early April | Start AI meeting capture from command palette | Business and above | Low | | Workers for Notion Agents | Early April (beta) | Sandboxed code execution for custom AI agent tools | Enterprise only | High | | AI-assisted Dashboard view | April 2 (rollout) | AI suggests charts and KPI cards from database properties | Plus and above | Medium | | Smart filters with relative values | Early April | "me" and relative date filters in views and API | All plans | Medium |
Release Timeline
Voice Input for AI Prompts
Notion added voice dictation scoped specifically to AI prompts on macOS and Windows desktop. Hold a configurable key, speak your prompt, and the transcription feeds directly into any AI action: inline AI, sidebar chat, or meeting notes.
The transcription runs through Notion's own speech pipeline rather than system-level dictation. Latency sits under one second for English prompts of typical length. Technical vocabulary works reasonably well, though acronyms and proprietary product names sometimes need manual correction.
When to Use Voice vs. Typing
Voice input works best for complex multi-step prompts. Saying "summarize the Q2 marketing plan and pull out the three riskiest assumptions, then format as a table" takes five seconds by voice versus thirty seconds of typing. For short prompts like "fix grammar" or "make this shorter", typing remains faster.
This is distinct from macOS system dictation (press Fn twice), which works in any text field for general text entry. Voice input is AI-prompt-only.
Workers for Notion Agents (Beta)
Workers are the most technically significant Notion AI release this month. They give developers a sandboxed code execution environment where custom tools run that Notion's AI agents can invoke during conversations.
How Workers Fit into the Agent Architecture
Workers function like function calling for Notion's AI. You write server-side code, register it with an agent, and the agent invokes it when a user's prompt matches the tool's description. Workers run in a sandboxed environment managed by Notion with no self-provisioned infrastructure required.
Worker Constraints
| Constraint | Current Limit | Practical Effect | |---|---|---| | Execution timeout | 30 seconds | Rules out long-running data processing | | Memory ceiling | 128 MB | Fine for API calls, tight for large dataset transforms | | Persistent storage | None | Every invocation starts from scratch | | Outbound networking | Domain allowlist | Cannot call arbitrary external APIs without approval | | Plan requirement | Enterprise only | Not available on Plus or Business plans |
What Workers Enable Today
- Pull records from CRM, analytics, or internal tools into Notion pages on demand
- Push data to Slack, email, or project management tools when an agent determines relevance
- Convert between formats (CSV to database entries, JSON to formatted pages)
- Run calculations and aggregations beyond Notion formula limits
AI-Assisted Dashboard View
The Dashboard view (part of the Notion 3.4 rollout) includes AI assistance for initial setup. When you create a Dashboard view, Notion AI analyzes database properties and suggests chart types, KPI cards, and metric groupings.
AI Suggestions vs. Manual Dashboard Setup
| Aspect | AI-Suggested | Manual Setup | |---|---|---| | Time to first view | 5 to 10 seconds | 10 to 30 minutes | | Accuracy | Good for standard patterns (status distribution, date trends) | Full control over every element | | Available chart types | Bar, line, donut | Same options | | KPI selection | Auto-picks numeric properties | You decide what matters | | Post-setup editing | Fully editable | Built from scratch |
The AI suggestions work best with databases that have clear numeric properties and status/select fields. For databases with only text and relation properties, expect less useful initial suggestions.
Performance note
Adding more than 6 to 8 chart widgets to a single Dashboard view slows page load noticeably. Keep dashboards focused on one team or metric domain rather than consolidating everything into one view.
Custom Meeting Note Instructions
Custom instructions let you define persistent templates for how Notion AI structures meeting summaries. You configure them at the workspace level (Settings, then Notion AI, then Meeting Notes):
- Tone: formal, conversational, or bullet-point only
- Sections: decisions, action items, open questions, parking lot items
- Length: brief one-pager or detailed transcript summary
- Team context: e.g., "weekly engineering standup, focus on blockers"
Once set, every new meeting note inherits these preferences. Individual users can override at the page level. The Cmd+K integration means you can start capturing meeting notes from anywhere in Notion without navigating to a specific page first.
Smart Filters with Relative Values
Smart filters add two capabilities that improve how AI features interact with database views:
- "me" as a people value: resolves to the authenticated user, no hardcoded user IDs needed
- Relative dates: "today", "tomorrow", "last_week", "this_week"
These work in both the Notion UI and the API. The AI significance: Notion's agents can now create personalized views ("My tasks due this week") with a single filter configuration that works for any team member, rather than building separate filters per user.
What These Releases Mean for Teams
The April 2026 Notion AI releases follow a clear pattern: moving AI from content generation toward workflow infrastructure. Voice input reduces friction for using AI features. Workers let AI agents reach outside Notion. Smart filters let agents build personalized views. Dashboard AI reduces setup time for data visualization.
For teams evaluating whether to invest more heavily in Notion's AI ecosystem, Workers are the signal to watch. If Notion loosens the current constraints (30-second timeout, 128 MB memory, domain allowlists) and extends Workers beyond Enterprise plans, it becomes a meaningfully different platform than it was six months ago.
What Is Still Missing
Related Reading
For detailed technical coverage of each feature, see our Notion AI release notes for April 2026. If you are tracking all Notion updates this month beyond AI, we cover the full changelog. For the API-specific changes, including Views API endpoints and smart filter parameters, see our developer guide.
If you are building AI agents that work across applications rather than within a single tool, Fazm is an open source macOS AI agent that automates desktop workflows by reading your screen and acting across any app. It handles the cross-application automation that Notion Workers cannot reach.
Fazm is an open source macOS AI agent. Open source on GitHub.