What can I do about this clause?

{'method': 'IN_APP', 'recipient': 'https://claude.ai/settings', 'difficulty': 'EASY', 'action_type': 'OPT_OUT_ARBITRATION', 'instructions': 'Log in to Claude.ai, navigate to Account Settings, and locate the model training or data usage preference toggle. Disable the option to use your conversations for model training. Note that this opt-out does not apply to feedback you provide or to content flagged for safety review.', 'deadline_days': None, 'deadline_hours': None}

Which regulatory agencies enforce this type of clause?

{'agency': 'FTC', 'reason': 'The FTC has authority under Section 5 of the FTC Act over unfair or deceptive practices, including misleading opt-out mechanisms for AI training data use.', 'complaint_url': 'https://reportfraud.ftc.gov/'}, {'agency': 'State_AG', 'reason': "California's Privacy Protection Agency and AG enforce CCPA opt-out rights; other state AGs may enforce analogous state consumer protection statutes regarding training data disclosures.", 'complaint_url': 'https://www.naag.org/find-my-ag/'}

What is the severity of this clause?

ConductAtlas classifies this Training Data Use and Opt-Out Carve-Outs clause as high severity. Severity reflects the magnitude of rights affected, the breadth of users impacted, and the degree of discretion the platform retains.

Training Data Use and Opt-Out Carve-Outs — Anthropic

Share 𝕏 Share in Share

What it is

Anthropic can use your Claude conversations to train its AI by default, and you can turn this off in settings — but even if you do, clicking thumbs up or down on a response, or having any message flagged for safety, means that content still gets used for training.

Consumer impact (what this means for users)

If you opt out of model training but ever rate a Claude response or have a conversation flagged for safety review, your conversation data is still used to train Anthropic's AI — limiting the practical effectiveness of the opt-out for most active users.

What you can do

⚠️ These actions may provide transparency or partial mitigation but may not fully address the underlying issue. Effectiveness varies by jurisdiction and individual circumstances.

Opt Out of Arbitration

Log in to Claude.ai, navigate to Account Settings, and locate the model training or data usage preference toggle. Disable the option to use your conversations for model training. Note that this opt-out does not apply to feedback you provide or to content flagged for safety review.

Cross-platform context

See how other platforms handle Training Data Use and Opt-Out Carve-Outs and similar clauses.

Compare across platforms →

Need full compliance memos? See Professional →

Why it matters (compliance & risk perspective)

The opt-out is partially illusory: two common and difficult-to-avoid actions (giving feedback and having content safety-reviewed) permanently override your opt-out preference, meaning many users who believe they have opted out may still be contributing training data.

View original clause language

We may use Materials to provide, maintain, and improve the Services and to develop other products and services, including training our models, unless you opt out of training through your account settings. Even if you opt out, we will use Materials for model training when: (1) you provide Feedback to us regarding any Materials, or (2) your Materials are flagged for safety review to improve our ability to detect harmful content, enforce our policies, or advance our safety research.

Institutional analysis (Compliance & legal intelligence)

(1) REGULATORY FRAMEWORK: This provision implicates GDPR Art. 6(1)(a) (consent as lawful basis), Art. 7 (conditions for consent including the right to withdraw), and Art. 5(1)(b) (purpose limitation) as enforced by EU/EEA Data Protection Authorities. It also engages CCPA §1798.120 (right to opt out of sale/sharing of personal information) and §1798.100 (right to know), enforced by the California Privacy Protection Agency (CPPA) and California AG. UK GDPR and the Data Protection Act 2018 apply to UK users, enforced by the ICO. The FTC Act Section 5 is relevant to whether the opt-out mechanism constitutes a deceptive practice given the scope of carve-outs. (2)

🔒

Compliance intelligence locked

Regulatory citations, enforcement risk, and due diligence action items.

Watcher $9.99/mo Professional $149/mo

Watcher: regulatory citations. Professional: full compliance memo.

Applicable agencies

FTC

The FTC has authority under Section 5 of the FTC Act over unfair or deceptive practices, including misleading opt-out mechanisms for AI training data use.
File a complaint →
State AG

California's Privacy Protection Agency and AG enforce CCPA opt-out rights; other state AGs may enforce analogous state consumer protection statutes regarding training data disclosures.
File a complaint →

Provision details

Document information

Document

Claude.ai Terms of Service

Entity

Anthropic

Document last updated

March 24, 2026

Tracking information

First tracked

March 6, 2026

Last verified

April 27, 2026

Record ID

CA-P-003163

Document ID

CA-D-00011

Evidence Provenance

Source URL

https://www.anthropic.com/legal/consumer-terms

Wayback Machine

View archived versions →

SHA-256

e757437a9d05ea816b5c1cddd3974f9a2ff93619333e14be4d368d9698b1e93f

Verified

✓ Snapshot stored ✓ Change verified

How to Cite

ConductAtlas Policy Archive
Entity: Anthropic | Document: Claude.ai Terms of Service | Record: CA-P-003163
Captured: 2026-03-06 19:30:31 UTC | SHA-256: e757437a9d05ea81…
URL: https://conductatlas.com/platform/anthropic/claudeai-terms-of-service/training-data-use-and-opt-out-carve-outs/
Accessed: April 29, 2026

Classification

Severity

High

Training Data Use and Opt-Out Carve-Outs