Compare acceptable use restrictions governance provisions between OpenAI and Anthropic. Provisions are extracted from monitored governance documents and classified by severity.
This provision establishes use restrictions that apply irrespective of how users attempt to direct model behavior, creating binding operational boundaries on the service. The hardcoded nature of these prohibitions means they operate as non-negotiable constraints on permitted applications rather than guidelines subject to contextual judgment.
Consumer impact
Users are prohibited from directing OpenAI models toward these specified use cases, and the restriction applies regardless of framing or instruction methods employed. The terms establish that these categories of use are not authorized under any circumstances of service use.
Opt-out available
No opt-out available
Actual clause text
Regardless of operator and user instructions, OpenAI prohibits use of its models or products for: generating CSAM or detailed sexual content involving minors; providing serious uplift to those seeking to create biological, chemical, nuclear, or radiological weapons with potential for mass casualties; creating cyberweapons or malicious code that could cause significant damage if deployed; generating content designed to facilitate actual (not fictional) violence against specific real people or groups; and taking actions that meaningfully undermine the ability of legitimate principals to oversee and correct advanced AI models.
AI-extracted from source document. Verify against original for legal use.
This provision establishes Anthropic's operational boundaries and compliance framework regarding content involving minors. The mandatory reporting mechanism creates a procedural obligation that integrates law enforcement notification into the service's governance structure when specified conduct is detected.
Consumer impact
Users operating under this policy are bound by a prohibition on specified categories of conduct involving minors, with the understanding that Anthropic maintains detection and reporting protocols. The provision clarifies that Anthropic defines minors as individuals under 18 years old regardless of jurisdiction, establishing uniform applicability across service usage.
Opt-out available
No opt-out available
Actual clause text
This includes using our products or services to: Create, distribute, or promote child sexual abuse material ("CSAM"), including AI-generated CSAM; Facilitate the trafficking, sextortion, or any other form of exploitation of a minor; Facilitate minor grooming, including generating content designed to impersonate a minor; Facilitate child abuse of any form, including instructions for how to conceal abuse; Promote or facilitate pedophilic relationships, including via roleplay with the model; Fetishize or sexualize minors, including in fictional settings or via roleplay with the model. Note: We define a minor or child to be any individual under the age of 18 years old, regardless of jurisdiction. When we detect CSAM (including AI-generated CSAM), or coercion or enticement of a minor to engage in sexual activities, we will report to relevant authorities.
AI-extracted from source document. Verify against original for legal use.
The Department of Defense designated Anthropic a supply chain risk after the company refused to rem…
AI Difference AnalysisCompliance
Stripe's arbitration clause is narrower than Amazon's in one key respect: it includes a small claims court carve-out that Amazon's clause does not. PayPal's clause is the most aggressive of the three, explicitly waiving jury trial rights in addition to class action rights. From a compliance perspective, Amazon presents the lowest risk for B2B contracts while PayPal creates the highest exposure for consumer-facing applications subject to CFPB oversight.