Cohere · Cohere Usage Policy · View original document ↗

Categorical Prohibition on Harmful Use Cases

High severity Medium confidence Explicitdocumentlanguage Unique · 0 of 325 platforms
Share 𝕏 Share in Share 🔒 PDF
Monitor governance changes for Cohere Create a free account to receive the weekly governance digest and monitor one platform for governance changes.
Create free account No credit card required.
Document Record

What it is

The policy lists specific types of applications and outputs that developers are never permitted to build or generate using Cohere's models, including content facilitating violence, hate speech targeting protected groups, fraud, and privacy violations.

This analysis describes what Cohere'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 establishes the outer boundary of permissible use and, because it is framed as a strict prohibition, violation would constitute a per se breach of the Terms of Service regardless of other approvals obtained.

Interpretive note: Some prohibited categories such as 'privacy violations' and 'disinformation' require contextual interpretation and may not be self-evidently applicable in all edge cases.

Consumer impact (what this means for users)

End users of Cohere-powered applications are protected by these prohibitions, which bar developers from using the models to generate content that facilitates violence against individuals, enables fraud targeting consumers, or produces material that violates personal privacy.

How other platforms handle this

Mistral AI Medium

Customer will not, and will not permit any other person (including any End User) to: ... (d) attempt to reverse engineer, decompile, or otherwise attempt to discover the source code or underlying components (e.g., algorithms, weights, or systems) of the Mistral AI Products, including using the Outpu...

23andMe Medium

By using the Services or creating an account, you represent, warrant and agree that: You are not an insurance company or an employer; and You will not use the Services for any investigative forensic genealogy uses.

Delta Airlines Medium

All content on this Internet site ("the delta.com website") is owned or controlled by Delta Air Lines and is protected by worldwide copyright laws.

See all platforms with this clause type →

Monitoring

Cohere has changed this document before.

Receive same-day alerts, structured change summaries, and monitoring for up to 10 platforms.

Start Watcher free trial Or create a free account →
▸ View Original Clause Language DOCUMENT RECORD
"
Certain use cases, such as violence, hate speech, fraud, and privacy violations, are strictly prohibited.

— Excerpt from Cohere's Cohere Usage Policy

ConductAtlas Analysis

Institutional analysis (Compliance & governance intelligence)

(1) REGULATORY LANDSCAPE: The prohibition on fraud-enabling applications engages with the FTC Act's prohibition on unfair or deceptive acts or practices. The prohibition on privacy violations engages with GDPR, CCPA, and sector-specific privacy laws. The prohibition on content targeting minors inappropriately engages with COPPA. The EU AI Act explicitly prohibits certain AI applications including manipulation of vulnerable groups and real-time biometric surveillance, which aligns with several of the prohibited categories in this provision. (2) GOVERNANCE EXPOSURE: Medium. The categorical nature of these prohibitions creates clear contractual liability for developers who deploy within prohibited categories. However, some prohibited categories, such as 'privacy violations' and 'disinformation,' involve factual and contextual determinations that may not always be clear-cut, creating interpretive uncertainty in edge cases. (3) JURISDICTION FLAGS: EU developers face the highest exposure given the EU AI Act's overlapping prohibitions. Illinois developers building applications involving biometric data should consider Illinois BIPA. Health-related prohibited uses engage HIPAA where covered entities or business associates are involved. (4) CONTRACT AND VENDOR IMPLICATIONS: Organizations that license or resell Cohere-powered capabilities to downstream customers should include equivalent prohibited use restrictions in their own customer agreements to maintain compliance with this provision and avoid upstream liability. (5) COMPLIANCE CONSIDERATIONS: Legal teams should map the prohibited use categories against their product roadmap and any third-party integrations to identify potential conflicts. Products in ambiguous categories should be reviewed against the special-approval list before deployment.

Full compliance analysis

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

Track 1 platform — free Try Watcher free for 14 days

Free: track 1 platform + weekly digest. Watcher: 10 platforms + same-day alerts. No credit card required.

Applicable agencies

  • FTC
    The FTC has enforcement authority over fraud and deceptive practices enabled by AI systems, directly relevant to the fraud and disinformation prohibition categories.
    File a complaint →

Applicable regulations

CFAA
United States Federal
Trump Executive Order on AI Policy Framework
US

Provision details

Document information
Document
Cohere Usage Policy
Entity
Cohere
Document last updated
May 5, 2026
Tracking information
First tracked
April 30, 2026
Last verified
May 12, 2026
Record ID
CA-P-011003
Document ID
CA-D-00442
Evidence Provenance
Source URL
Wayback Machine
Content hash (SHA-256)
2937f674a79ab03784eab9a8774b7c807068d6f695cd81b3eb7bc9419a338c76
Analysis generated
April 30, 2026 06:46 UTC
Methodology
Evidence
✓ Snapshot stored   ✓ Hash verified
Citation Record
Entity: Cohere
Document: Cohere Usage Policy
Record ID: CA-P-011003
Captured: 2026-04-30 06:46:20 UTC
SHA-256: 2937f674a79ab037…
URL: https://conductatlas.com/platform/cohere/cohere-usage-policy/categorical-prohibition-on-harmful-use-cases/
Accessed: May 13, 2026
Permanent archival reference. Stable identifier suitable for legal filings, compliance documentation, and research citation.
Classification
Severity
High
Categories

Other risks in this policy

Professional Governance Intelligence

Need to monitor specific governance provisions?

Professional includes provision-level monitoring, governance timelines, regulatory mapping, and audit-ready analysis.

Arbitration clauses AI governance Data rights Indemnification Retention policies
Start Professional free trial

Or start with Watcher →

Built from archived source documents, structured governance mappings, and historical version tracking.

Frequently Asked Questions

What does Cohere's Categorical Prohibition on Harmful Use Cases clause do?

This provision establishes the outer boundary of permissible use and, because it is framed as a strict prohibition, violation would constitute a per se breach of the Terms of Service regardless of other approvals obtained.

How does this clause affect you?

End users of Cohere-powered applications are protected by these prohibitions, which bar developers from using the models to generate content that facilitates violence against individuals, enables fraud targeting consumers, or produces material that violates personal privacy.

Is ConductAtlas affiliated with Cohere?

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