ClauseLens

Compare / Azure OpenAI vs Microsoft Copilot

Azure OpenAI vs Microsoft Copilot: which keeps your data private?

Side by side, from each vendor’s own documents. Open either tool for the exact quotes behind every cell.

Trains on your data
Does not train
  • FreeTrains unless you opt out
  • Pro / PlusTrains unless you opt out
  • EnterpriseDoes not train
How to opt out
Not stated
Easy opt-out offered; method not stated in document
Data retention
30 days for abuse monitoring (per community Q&A; primary doc does not state)
Configurable by admins via Microsoft Purview retention policies
Where data is stored
Processed in customer-specified geography; Global/DataZone deployments may process elsewhere
EU Data Boundary for EU customers; others may be processed in US, EU, or other regions
DPA available
Yes
Yes
Subprocessors
Not stated
Anthropic, OpenAI (AI model providers); Microsoft-hosted models
Certifications claimed
Azure-wide: ISO/IEC 27701, ISO/IEC 27018, HIPAA, HITRUST (Azure-specific for Azure OpenAI not stated)
GDPR, ISO 27001, HIPAA, ISO 42001
Human review of your data
Humans may review
No human review

Azure OpenAI on training

“The models are stateless: no prompts or completions are stored in the model. Additionally, prompts and completions are not used to train, retrain, or improve the base models.”

checked Nov 16, 2026 · all quotes

Microsoft Copilot on training

“Prompts, responses, and data accessed through Microsoft Graph aren't used to train foundation LLMs, including those used by Microsoft Copilot.”

checked Nov 9, 2026 · all quotes

Informational, not legal advice. Every verdict is what the vendor’s own documents state, quoted word for word. Read the cited source before you decide, and check your own contract: negotiated terms can differ.