Provision record
Stability AI · Stability AI Terms of Service · View original document ↗

Input and Output Use for Model Training

High severity Medium confidence Explicit document language Unique · 0 of 352 platforms
Stay ahead of the changes
Track Stability AI and get the diff the day its terms change.
Share 𝕏 Share in Share 🔒 PDF
Document Record

What it is

Stability may use submitted Inputs and Outputs to improve and develop its services, including training its models, unless users opt out through the mechanism referenced in Section 4c.

This analysis describes what Stability AI's agreement states, permits, or reserves. It does not constitute a legal determination about enforceability. Regulatory applicability and practical outcomes may vary by jurisdiction, enforcement context, and individual circumstances. Read our methodology

ConductAtlas Analysis

Why it matters (compliance & governance perspective)

This provision authorizes the use of user-submitted content, including text, images, and other inputs, for model training and service development. An opt-out mechanism is available but requires affirmative user action; the default position under the terms is that Inputs and Outputs may be used for training.

Interpretive note: The lawful basis for processing Inputs and Outputs for model training under GDPR is not specified in this document; the Privacy Policy governs data processing details and would need to be reviewed to assess full regulatory exposure.

Recent Activity

This document changed recently

Medium Sep 1, 2026

The updated terms establish a lower age threshold for audio services (13+) but require parental consent and active supervision for users under 18. All other services remain restricted to users 18 and older. The company expanded its license to use and process submitted content for service provision and maintenance. The arbitration clause now applies only to US users; Canadian users are no longer subject to mandatory arbitration. Use restrictions were clarified to prohibit developing competitive AI models, selling or sublicensing the Services, and removing AI-generated content markings, with limited exceptions expressly approved in writing.

View change record →

Clause Stability Stable

0
Changes
5
Months Monitored
Jul 9, 2026
First Seen
Jul 9, 2026
Last Seen

Consumer impact (what this means for users)

Under this clause, submitted Inputs and Outputs are available for Stability's use in model training and service improvement by default. Users who do not affirmatively opt out through the mechanism linked in Section 4c operate under terms that permit this use of their submitted content.

Cross-platform context

See how other platforms handle Input and Output Use for Model Training and similar clauses.

Compare across platforms →
▸ View Original Clause Language DOCUMENT RECORD
"
As described in our Privacy Policy, we may use Inputs and Outputs to improve and develop our Services (but you can opt-out to prevent us from using your Inputs and Outputs to train our models, learn more), comply with applicable law, enforce our terms and policies, and to keep our Services safe.

Excerpt from Stability AI's Terms of Service

ConductAtlas Analysis

Institutional analysis (regulatory & governance intelligence)

1.

Insight

Unlock the full institutional analysis

Enforcement risk, jurisdiction flags, contract triggers, and due diligence action items.

Applicable agencies

  • Federal Trade Commission (ftc)
    Oversees unfair or deceptive business practices and can investigate companies that mislead consumers about data collection, sharing, or use.
    Who can file: Anyone affected by the company's practices (US or international)
    What you need: Your account details, a timeline of relevant events, and a description of the specific issue
    What to expect: Complaints inform FTC enforcement priorities and investigations but do not result in individual resolution or compensation
    File a complaint →
  • State Attorney General
    State AGs in California, New York, Texas, and other states can investigate violations of state consumer protection and privacy laws, including CCPA (California), SHIELD Act (New York), and equivalents.
    Who can file: Residents of states with comprehensive privacy laws — primarily California, Virginia, Colorado, Connecticut, and Utah
    What you need: Evidence of the violation, explanation of how your state rights were affected, and your account or contact information with the company
    What to expect: Outcomes vary by state. May result in investigation, enforcement action, or requirement for the company to change practices. No direct individual compensation in most cases.

    Search "[your state] attorney general consumer complaint" to find your state's direct complaint form

Provision details

Document information
Document
Stability AI Terms of Service
Entity
Stability AI
Document last updated
May 5, 2026
Tracking information
First tracked
April 28, 2026
Last verified
July 9, 2026
Record ID
CA-P-015337
Document ID
CA-D-00329
Evidence Provenance
Source URL
Wayback Machine
Content hash (SHA-256)
98789d636c554bd7467c35eb18589d01a6c721c0b90ef04a688c4183b8a8dee9
Analysis generated
April 28, 2026 05:27 UTC
Methodology
Evidence
✓ Snapshot stored   ✓ Hash verified
Citation Record
Entity: Stability AI
Document: Stability AI Terms of Service
Record ID: CA-P-015337
Captured: 2026-04-28 05:27:04 UTC
SHA-256: 98789d636c554bd7…
URL: https://conductatlas.com/platform/stability-ai/stability-ai-terms-of-service/provision/CA-P-015337/input-and-output-use-for-model-training/
Accessed: Sept. 8, 2026
Permanent archival reference. Stable identifier suitable for legal filings, compliance documentation, and research citation.
Classification
Severity
High
Categories

Other risks in this policy

Get the research letter

Companies change their terms quietly. We read every version and catch what actually changed. One email a week on the changes that matter and what they mean.

Frequently Asked Questions

What does Stability AI's Input and Output Use for Model Training clause do?

This provision authorizes the use of user-submitted content, including text, images, and other inputs, for model training and service development. An opt-out mechanism is available but requires affirmative user action; the default position under the terms is that Inputs and Outputs may be used for training.

How does this clause affect you?

Under this clause, submitted Inputs and Outputs are available for Stability's use in model training and service improvement by default. Users who do not affirmatively opt out through the mechanism linked in Section 4c operate under terms that permit this use of their submitted content.

Is ConductAtlas affiliated with Stability AI?

No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Stability AI.