ModelCharter

Is Midjourney safe for work?

High risk · 97

Midjourney, Inc. · Image · facts partly unverified, check sources

Midjourney is high-risk for default at-work use (97/100): it trains on your data by default.

Watch out: All prompts and generated images are public by default and licensed for model training, and Stealth Mode (Pro/Mega only) merely hides outputs from the public gallery without exempting them from training, so confidential work-related content should not be used.

Data and compliance facts

Trains on consumer-tier data
Yes
Trains on business-tier data
Yes
Training opt-out available
No
SOC 2 Type II
Unverified
ISO 27001
Unverified
ISO 42001 (AI management)
Unverified
GDPR Data Processing Addendum
Unverified
HIPAA BAA
Unverified
EU data residency
Unverified
SSO / SAML
Unverified
Data retention
Not specified in Midjourney's published Terms of Service or help docs; no defined retention or deletion timeline was found, and Midjourney retains a perpetual license to submitted content.

Why it scores 97 out of 100

  • +30Trains on your data by default. On the free/consumer tier your inputs are used to train the vendor's models unless you change a setting.
  • +22Trains on business-tier data. Even paid/team data may be used for training. That is unusual and high-risk.
  • +12No SOC 2 Type II. No independent SOC 2 Type II attestation found.
  • +6No ISO 27001. No ISO/IEC 27001 information-security certification found.
  • +12No GDPR DPA. No standard Data Processing Addendum offered, a problem for any team with EU/UK users.
  • +6No EU data residency. Data cannot be guaranteed to stay in the EU.
  • +4No SSO/SAML. No enterprise single sign-on, so account access is harder to govern.
  • +5No HIPAA BAA. No Business Associate Agreement, so do not use it with protected health information.