# Microsoft 365 Copilot — AI Agents & Assistants

Microsoft 365 Copilot by Microsoft Corporation holds a panel rating of 3/5 from 6 AI judges (category median 3/5), as of 2026-10-28. Provenance: Rest of world; sovereignty: not determined.

- Vendor: Microsoft Corporation (https://learn.microsoft.com/en-us/microsoft-365/copilot/copilot-overview)
- Category: [AI Agents & Assistants](https://whats-best.ai/software/ai-agents.md)
- Provenance (editorial classification): Rest of world
- Canonical page: https://whats-best.ai/software/ai-agents/microsoft-365-copilot
- Alternatives: https://whats-best.ai/software/ai-agents/microsoft-365-copilot/alternatives ([markdown](https://whats-best.ai/software/ai-agents/microsoft-365-copilot/alternatives.md))
- Data as of: 2026-10-29 · evaluated 2026-10-28 · oldest source captured 2026-10-26

**AI-generated.** Every score, rationale and verdict here is written by a panel of AI judges from public evidence — an analysis, not a hands-on test. Every fact quotes the vendor page it was read from. No score can be bought. How scores are made: https://whats-best.ai/methodology

## At a glance

- Vendor: Microsoft Corporation
- Headquarters: Not stated on the vendor’s pages as of 26 Oct 2026
- Data hosting: Not stated on the vendor’s pages as of 26 Oct 2026
- Free plan: Not stated on the vendor’s pages as of 26 Oct 2026

## The panel's verdict

Microsoft 365 Copilot is an AI assistant add-on on top of Microsoft 365, and the bench scored it strongest on data use and training and on integrations and knowledge, both at 6-8: judges cited documented commitments that prompts, responses and Graph-accessed data are not used to train foundation models, a downloadable data protection addendum, ISO 27001 and ISO 42001 certifications, and a connectors API whose grounded answers store citations to their sources. Sovereignty is where the judges genuinely split, spanning 4-6: the works council member credited the EU Data Boundary commitment in the Product Terms since March 2024 and a downloadable Data Protection Addendum, while the skeptic, head of IT and data protection officer weighed Anthropic's models being excluded from that boundary, the United States named among storage locations, and a US contracting vendor. Scores on processing and models span 5-6 on the same fault line, and agent capabilities 5-6, with documented jailbreak and prompt-injection classifiers but no public information on human approval before consequential actions. Pricing transparency scores sit at 2: no captured page states a price.

**Speaks for it:**
- Documents that prompts, responses and Graph-accessed data are not used to train foundation models, backed by a downloadable data protection addendum and ISO 27001 and ISO 42001 certifications.
- Grounding surfaces only data a user has at least view permission for, with the Semantic Index honoring the identity-based access boundary.
- A documented connectors API with a five-step build, SDKs and Teams bots brings external data into Microsoft Graph, and grounded answers store citations to their sources.
- Users can delete their own Copilot activity history through the My Account portal, and admins manage stored prompts and responses through Purview and Content search.
- Guardrails are documented openly — jailbreak and cross-prompt injection attack classifiers, content harm filters, and protected-material detection.

**Held against it:**
- The captured pages give different statements on training use: the Copilot privacy documentation says prompts, responses and Graph-accessed data are not used to train foundation models, while the general privacy statement says Microsoft "may use your data to develop and train our AI models".
- Anthropic's models are documented as currently excluded from the EU Data Boundary, and the privacy statement names the United States among storage locations.
- We found no public information on human approval before consequential agent actions, run logs, or triggers and schedules.
- We found no public information on stated retention periods or zero-data-retention terms with the named model providers, OpenAI and Anthropic.
- We found no public information on SSO with SCIM provisioning, group-level permissions per agent or connector, or exportable audit logs.

**Best for:**
- You already run Microsoft 365 and want answers that respect each user's existing view permissions.
- You need to ground answers in external systems through a documented connectors API, with citations stored with each grounded response.
- Your procurement requires a downloadable data protection addendum and certifications such as ISO 27001 and ISO 42001.
- Your works-council review turns on pseudonymised system-generated logs and users' ability to delete their own Copilot activity history.

**Avoid if:**
- You need every model behind the product to process inside the EU — Anthropic's models are documented as currently excluded from the EU Data Boundary.
- Your tenant uses Microsoft 365 Multi-Geo — ask the vendor: the public pages we read do not show it
- You are a small business unwilling to accept an annual commitment — the Microsoft Copilot Business add-on is documented with an annual commitment, up to 300 licensed users per tenant.

## Scores by judge and criterion

Scores out of 5, each judge scoring independently against the published rubric (v1). "n/r" = not rated: the vendor publishes no price, or the judge found no public information on that criterion.

| Criterion | The Head of IT | The Data Protection Officer | The Automation Lead | The Customer-Service Director | The Works Council Member | The Skeptic |
| --- | --- | --- | --- | --- | --- | --- |
| Data use, training & retention | 3.5/5 | 3/5 | 3.5/5 | 3.5/5 | 4/5 | 3.5/5 |
| Models & processing location | 3/5 | 2.5/5 | 3/5 | 2.5/5 | 3/5 | 2.5/5 |
| Admin, identity & access controls | 3.5/5 | 3/5 | 3.5/5 | 3.5/5 | 3/5 | 3/5 |
| Integrations, connectors & knowledge | 3.5/5 | 3/5 | 3.5/5 | 3.5/5 | 4/5 | 3.5/5 |
| Agent capabilities & guardrails | 3/5 | 2.5/5 | 3/5 | 2.5/5 | 2.5/5 | 2.5/5 |
| European sovereignty | 2/5 | 2/5 | 2.5/5 | 2.5/5 | 3/5 | 2/5 |
| Pricing transparency | n/r | n/r | n/r | n/r | n/r | n/r |
| **Weighted verdict** | **3/5** | **2.5/5** | **3/5** | **3/5** | **3/5** | **3/5** |

### The judges

- **The Head of IT** (v1.0): Has to roll an AI assistant out to eight hundred staff without a second shadow-IT wave. Wants SSO and SCIM, groups that decide who gets which model, agent and connector, answers that respect SharePoint permissions, audit logs for the SIEM, and a deployment option that is not "trust us".
- **The Data Protection Officer** (v1.0): Signs the DPA and gets asked where a prompt goes. Wants every model provider named as a subprocessor with its region, training excluded for all of them, retention the company controls, and an EU-only switch that also covers agents and connectors — not just the chat window.
- **The Automation Lead** (v1.0): Builds the agents that file tickets, draft offers and chase approvals across the ERP, CRM and wiki. Wants tools and actions with scoped permissions, MCP and an API, approval steps before anything is sent, run logs, and a way to test an agent before the whole company uses it.
- **The Customer-Service Director** (v1.0): Runs a contact centre with forty agents and wants half the routine calls and chats answered by an AI agent. Wants handover to a human with the full context, actions in the order and ticket systems, evaluation before go-live, and a price per conversation the CFO can forecast.
- **The Works Council Member** (v1.0): Has a co-determination right the moment a tool can monitor staff, and an AI assistant logs every prompt. Wants usage reports that can be aggregated or pseudonymised, prompt logging rules set out in writing, a human approving what agents do to colleagues, and no training on staff input.
- **The Skeptic** (v1.0): Has read "European AI" on a wrapper that sends every prompt to a US model, "agents" that are saved prompts, and "we never train on your data" with no word on the model provider behind it. Asks which model answers, where it runs, what it may do unsupervised, and what a thousand seats cost.

## European sovereignty — proven facts

Built only from facts on the vendor's own pages; a dimension we could not prove is left open, never scored as zero. Headline: not determined.

| Dimension | Finding | Points | Source |
|---|---|---|---|
| Legal entity | Not determined | — | — |
| Ownership | Not determined | — | — |
| Data residency | Not determined | — | — |
| Subprocessors | Not determined | — | — |

## Key facts

Each fact is the vendor's own statement, cited to the source page it was read from ([S*n*] → Sources).

**Compliance**
- Abuse monitoring opt out: yes [S3]
- Certifications: GDPR · ISO 27001 · HIPAA · ISO 42001 [S3]
- Child consent age: 13 · yes · yes [S6]
- Cookie consent: Manage cookies in footer · yes [S6]
- DPA addendum status: Product Terms site (formerly OST) [S5]
- DPA download availability: yes [S5]
- DPA terms location: Data processing and security terms for Products subscribed under the Product Terms site are defined in the Microsoft Online Services Data Protection Addendum (DPA). [S5]
- Dpf membership: EU-U.S. Data Privacy Framework · UK Extension to the EU-U.S. Data Privacy Framework · Swiss-U.S. Data Privacy Framework [S6]
- Encryption at rest transit: BitLocker · per-file encryption · TLS · IPsec [S3]
- EU AI act: yes [S3]
- GDPR commitment: yes · yes [S3]
- International transfer safeguards: yes [S6]
- ISO standards: ISO/IEC 27018 [S3]
- Log pseudonymization required: All personal data in system-generated logs must be pseudonymized (encryption, masking, tokenization, data blurring) [S4]
- Optin data collection: yes [S6]
- Pseudonymization controls: Reidentification-data controls are the same as for Customer Data [S4]
- Purview encryption honored: yes [S3]

**Data**
- Admin tools: Content search · Microsoft Purview [S3]
- Retention minimization: Retention policies set at minimum required time per log type [S4]
- Stored interactions: user prompt · Copilot response · citations · yes [S3]
- Teams export API: yes [S3]
- User deletion: My Account portal · yes [S3]

**Hosting**
- Anthropic EU exclusion: yes [S3]
- Azure region configuration: Azure regional services in EU Data Boundary regions are in-scope; non-regional services require configuration; Azure Resource Manager must be configured to store Professional Services Data in boundary [S4]
- Data residency: Advanced Data Residency · Multi-Geo Capabilities · 2024-03-01 [S3]
- EU data boundary: yes [S3]
- Non EU processing regions: US · EU · other [S3]
- Storage regions: user's region · United States · other jurisdictions where Microsoft or affiliates/subsidiaries/service providers operate [S6]

**Integrations**
- API available: Copilot connectors API · yes [S8]
- Bicep templates: yes [S7]
- Code samples: yes [S7]
- Graph connectors: yes · yes [S3]
- No teams subscriptions: Teams-less Education or Business subscriptions eligible · yes [S2]
- Sdks: yes [S7]

**Legal**
- Statement last updated: September 2026 [S6]

**Pricing**
- Business addon: Microsoft Copilot Business · SMB · yes · yes · 300 [S2]
- Developer subscription: might qualify via Microsoft 365 Developer Program · yes [S7]
- Education availability: EES · CSP · yes · faculty, staff, and students aged 13+ [S2]
- Eligible plans office365: Office 365 E5 · Office 365 E3 · Office 365 E1 · Office 365 F3 [S2]

**Product**
- Ad personalization optout: yes [S6]
- Admin agent control: yes [S3]
- Agents toolkit channels: Microsoft 365 Agents Toolkit is a Visual Studio extension [S9]
- AI governance standards: Microsoft AI Principles · Microsoft Responsible AI Standard [S6]
- AI training use: yes [S6]
- Boundary countries: EU plus EFTA countries [S4]
- Change notifications: Microsoft Graph Change Notifications available (used with Azure Event Hubs) [S7]
- Community forum: yes [S9]
- Connector build steps: choose data sources · use Copilot connectors API · create connection and configure search/activity settings · define and register schema with semantic labels · ingest and manage external items including content access [S8]
- Content harm filters: Hate & Fairness · Sexual · Violence · Self-harm [S3]
- Continuing transfers: Some Customer Data, personal data, and Professional Services Data continues to be transferred outside the EU Data Boundary in limited circumstances [S4]
- Copilot connectors purpose: Bring external data into Microsoft Graph for Copilot and Microsoft Search [S8]
- Copilot studio builder: Copilot Studio builds agents and extends Microsoft 365 Copilot on a visual canvas [S9]
- Copyright commitment: customer used built-in guardrails and content filters · yes [S3]
- Customer data definition: Customer Data excludes Professional Services Data and configuration information such as technical settings and resource names [S4]
- Data rights metrics publication: yes [S6]
- Diagnostic tool: service plan diagnostic tool for per-user eligibility verification · yes [S2]
- Discord community: yes [S9]
- Dynamics power platform configuration: Dynamics 365/Power Platform require EFTA Macro Region Geography tenant plus billing address in EU Data Boundary country [S4]
- Enterprise terms precedence: yes [S6]
- EU data boundary scope: EU Data Boundary stores and processes Customer Data and personal data for Azure, Dynamics 365, Power Platform, Microsoft 365; Professional Services Data stored at rest [S4]
- Family safety tools: yes [S6]
- Foundry model count: foundation · open · task · industry-specific · 11000 [S9]
- Foundry platform: Microsoft Foundry designs, customizes and manages AI applications and agents at scale [S9]
- Foundry sdk: Microsoft Foundry SDK provides an integrated library of SDKs and APIs [S9]
- Government availability: add-on · GCC · GCC High · DoD [S2]
- Gpc signal: yes [S6]
- Graph explorer: yes [S7]
- Graph toolkit: Microsoft Graph Toolkit controls: login, people picker, person card [S7]
- Ground copilot: ground Copilot responses in external work content and context [S8]
- Last updated: Last updated: February 26, 2025 [S4]
- M365 copilot APIs: Microsoft 365 Copilot APIs enable generative AI using Microsoft 365 data with search, retrieval, chat, and compliance capabilities [S9]
- Mail API: Mail API and change notifications for processing and responding to email messages [S7]
- Microsoft365 configuration: Microsoft 365 in scope based on sign-up location in EU/EFTA [S4]
- Microsoft365 multi geo exclusion: Microsoft 365 Multi-Geo customers are not in scope [S4]
- Model auto upgrade: yes [S9]
- Model updates no setting change: yes [S3]
- Natural language search: natural language search over ingested content via Copilot and Copilot Search [S8]
- No training on customer data: no [S3]
- Optional connected experiences: yes [S3]
- Ownership of output: no [S3]
- Permission based grounding: only surfaces data users have at least view permissions for [S3]
- Powershell automation: PowerShell provisioning of users, groups, Teams; M365 usage reports; security alerts [S7]
- Privacy controls effect: Excel · OneNote · Outlook · PowerPoint · Word [S3]
- Privacy dashboard: yes [S6]
- Professional services definition: Professional Services = Microsoft consulting services and technical support services (e.g., Unified Support, Premier Support) [S4]
- Prompt injection mitigation: yes [S3]
- Protected material detection: yes [S3]
- Relevant results surfacing: surface external content alongside Microsoft 365 files and data [S8]
- Semantic index access boundary: yes [S3]
- Single endpoint: Access rich, people-centric data in the Microsoft Cloud via a single endpoint with the Microsoft Graph REST APIs and SDKs. [S7]
- Student copilot chat: Copilot Chat for students · yes · 13+ [S2]
- Teams bots: Teams bots for surveys, notifications, and required-action alerts [S7]
- Tenant isolation: Entra authorization + RBAC [S3]
- Third party signin data sharing: yes [S6]
- US state privacy notice: yes [S6]
- Vscode agent development: Agents can be created, tested and deployed from Visual Studio Code via AI Toolkit, GitHub Models, and Microsoft Foundry extensions [S9]
- Web search via bing: Bing Search service [S3]
- Windows statement scope: Windows 10 · Windows 11 [S6]

**Subprocessors**
- Anthropic: Anthropic · AI models in Copilot experiences [S3]
- Openai: OpenAI · AI models in Copilot experiences [S3]

**Support**
- Dev resources: connector build guide · quick start · code samples · interactive demo videos [S8]

## Questions

### Is Microsoft 365 Copilot European?

No. whats-best.ai classifies Microsoft 365 Copilot as Rest of world.

### Where does Microsoft 365 Copilot store data?

Not stated on the vendor’s pages as of 26 Oct 2026.

### Does Microsoft 365 Copilot have a free plan?

Not stated on the vendor’s pages as of 26 Oct 2026.

### What are the best alternatives to Microsoft 365 Copilot?

As of 28 Oct 2026, the panel rates [Langdock (4/5)](https://whats-best.ai/software/ai-agents/langdock), [Dust (3.5/5)](https://whats-best.ai/software/ai-agents/dust) and [Unique AI (3.5/5)](https://whats-best.ai/software/ai-agents/unique-ai) highest among the other 15 AI Agents & Assistants products evaluated. Ratings that look equal are rounded to half stars; the order uses the unrounded rating, then the name. The highest-rated European alternative is [Langdock](https://whats-best.ai/software/ai-agents/langdock) (EU-Made). Ratings are the AI panel’s opinion, formed from public evidence: [how scores are made](https://whats-best.ai/methodology#ordering). The full list, with European products first: [Microsoft 365 Copilot alternatives](https://whats-best.ai/software/ai-agents/microsoft-365-copilot/alternatives).

## Head-to-head

- [Microsoft 365 Copilot vs Gemini for Google Workspace](https://whats-best.ai/compare/gemini-for-google-workspace-vs-microsoft-365-copilot.md)

## What this page leaves out

- 4 pricing facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 3 compliance facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 2 product facts could not be confirmed on the vendor’s page as captured and were left out of this page and of the panel’s material.
- 1 hosting fact could not be confirmed on the vendor’s page as captured and was left out of this page and of the panel’s material.
- 1 sovereignty dimension could not be confirmed on the vendor’s own pages and is shown as unknown.

## Sources

- [S1] Product documentation (overview): https://learn.microsoft.com/en-us/microsoft-365/copilot/copilot-overview (checked 2026-10-26)
- [S2] Licensing and pricing documentation: https://learn.microsoft.com/en-us/microsoft-365/copilot/copilot-license-options (checked 2026-10-26)
- [S3] Data, privacy and security documentation: https://learn.microsoft.com/en-us/microsoft-365/copilot/microsoft-365-copilot-privacy (checked 2026-10-26)
- [S4] EU Data Boundary documentation: https://learn.microsoft.com/en-us/privacy/eudb/eu-data-boundary-learn (checked 2026-10-26)
- [S5] Data protection addendum: https://www.microsoft.com/licensing/docs/view/Microsoft-Products-and-Services-Data-Protection-Addendum-DPA (checked 2026-10-26)
- [S6] Privacy policy: https://www.microsoft.com/en-us/privacy/privacystatement (checked 2026-10-26)
- [S7] Integrations, connectors & knowledge — found from sitemap: https://developer.microsoft.com/en-us/graph/rest-api (checked 2026-10-26)
- [S8] Integrations, connectors & knowledge — found from sitemap: https://developer.microsoft.com/en-us/microsoft-365/copilot-connectors (checked 2026-10-26)
- [S9] Agent capabilities & guardrails — found from sitemap: https://developer.microsoft.com/en-us/agents (checked 2026-10-26)

---

Cite as: whats-best.ai, "Microsoft 365 Copilot — AI Agents & Assistants", AI panel analysis, as of 2026-10-29. https://whats-best.ai/software/ai-agents/microsoft-365-copilot

Corrections: https://whats-best.ai/correct/1036?target=page