Get the weekly 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. No account.
The policy prohibits using Cohere services in contexts where failure could cause death or serious harm, or where the service is used to make automated decisions affecting individuals in domains such as employment, healthcare, housing, financial services, or law enforcement. Back-office applications including document summarization, transcription, and internal knowledge agents are explicitly excluded from this prohibition.
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
This provision establishes a categorical restriction on deploying Cohere services for automated individual determinations in high-stakes domains, with direct operational implications for enterprise customers in HR, lending, insurance, healthcare, and public sector contexts. The back-office carve-out creates a meaningful but operationally imprecise boundary that compliance teams must evaluate against specific deployment architectures.
Interpretive note: The boundary between a prohibited automated determination and a permissible back-office use is not exhaustively defined, and the policy does not address human-in-the-loop configurations, creating interpretive uncertainty for complex enterprise deployments.
The updated policy removes all substantive acceptable use requirements that were previously posted and enforceable. Users no longer have a referenced standard defining what conduct is prohibited on the platform. The removal of enforcement procedures means users cannot verify what conduct may trigger access restriction, suspension, or termination. The elimination of the child safety and sexually explicit content prohibitions from the posted policy creates uncertainty about whether these protections remain in effect through other terms or have been abandoned.
View change record →This provision restricts the contexts in which Cohere-powered applications may be used to make or inform automated decisions about individuals in areas including employment, healthcare, housing, and financial services. Under this clause, consumers interacting with applications built on Cohere's services in these domains should not be subject to automated determinations generated by those services.
Cross-platform context
See how other platforms handle High Risk Activities Prohibition and similar clauses.
Compare across platforms →Monitoring
Cohere has changed this document before.
Receive same-day alerts, structured change summaries, and monitoring for up to 25 platforms.
"High Risk Activities. Activities (1) where the use or failure of the Cohere Services could reasonably be expected to result in death, harm to psychological or physical health or safety, or severe environmental or property damage; or (2) that use the Cohere Services for automated determinations about individuals in domains that affect their rights, safety, or access to essential services and benefits (e.g., employment, education, healthcare, migration, housing, law enforcement, legal advice/decisions, or financial or insurance products or services). For the avoidance of doubt, backoffice uses (e.g., document summarization, transcription, internal knowledge agents, etc.) are not considered High Risk Activities under this Usage Policy.Excerpt from Cohere's Usage Policy
1) REGULATORY LANDSCAPE: This provision directly engages the EU AI Act's classification of high-risk AI systems, which includes systems used in employment, education, healthcare, migration, housing, financial services, and law enforcement. The EU AI Act imposes conformity assessments, transparency requirements, and human oversight obligations on high-risk AI deployments; this policy prohibition may interact with those obligations by restricting Cohere's use in those contexts at the contractual level. The FTC has issued guidance on automated decision-making and unfair or deceptive practices that is also relevant. Enforcement authority includes EU national market surveillance authorities under the AI Act and the FTC in the US. 2) GOVERNANCE EXPOSURE: High. The prohibition on automated determinations affecting individual rights in employment, healthcare, housing, and financial services creates significant compliance exposure for enterprise customers who may be deploying Cohere in adjacent workflows. The boundary between a prohibited automated determination and a permissible back-office support tool is not exhaustively defined in the policy, creating interpretive uncertainty for complex enterprise implementations. 3) JURISDICTION FLAGS: EU and EEA deployments create heightened exposure due to the EU AI Act's specific high-risk system classifications, which align closely with the domains enumerated in this provision. US financial services and healthcare contexts engage sector-specific regulations including ECOA, FCRA, and HIPAA. Employment-related automated decision-making creates exposure in states with emerging algorithmic accountability laws, including New York City Local Law 144 on automated employment decision tools. 4) CONTRACT AND VENDOR IMPLICATIONS: Enterprise procurement teams should evaluate whether their specific Cohere deployment use cases fall within the prohibited automated determination category or the back-office carve-out, and document that analysis. Vendor agreements with Cohere should address the implications of access termination in the event of a disputed policy interpretation, particularly where Cohere services are integrated into critical business workflows. The policy does not define an appeals or dispute process for enforcement decisions. 5) COMPLIANCE CONSIDERATIONS: Legal and compliance teams should map each Cohere deployment use case against the enumerated high-risk domains and document the rationale for any classification as a back-office use. Organizations in financial services, healthcare, or HR technology should assess whether their implementation involves any automated output that directly informs individual determinations, even if a human reviews the output, as the policy language does not explicitly address human-in-the-loop configurations.
Full institutional analysis
Regulatory citations, enforcement risk, and due diligence action items.
Monitor: same-day alerts on the platforms you choose. Analyst: full institutional analysis.
Compliance Governance Intelligence
Need to monitor specific governance provisions?
Compliance includes provision-level monitoring, governance timelines, regulatory mapping, and audit-ready analysis.
Built from archived source documents, structured governance mappings, and historical version tracking.
This provision establishes a categorical restriction on deploying Cohere services for automated individual determinations in high-stakes domains, with direct operational implications for enterprise customers in HR, lending, insurance, healthcare, and public sector contexts. The back-office carve-out creates a meaningful but operationally imprecise boundary that compliance teams must evaluate against specific deployment architectures.
This provision restricts the contexts in which Cohere-powered applications may be used to make or inform automated decisions about individuals in areas including employment, healthcare, housing, and financial services. Under this clause, consumers interacting with applications built on Cohere's services in these domains should not be subject to automated determinations generated by those services.
No. ConductAtlas is an independent monitoring service. We are not affiliated with, endorsed by, or sponsored by Cohere.