ElevenLabs Announces FedRAMP 20x Class A Certification
The company says the certification covers ElevenAgents and speech APIs only under zero-retention and U.S.-residency settings, narrowing the path into non-sensitive federal workloads.
Edited by Tyronne Panaino
ElevenLabs announced on October 1 that it had achieved FedRAMP 20x Class A certification for three parts of its federal offering: ElevenAgents, Text to Speech and Speech to Text. The company says coverage applies when those services run in Zero Retention Mode with U.S. data residency, a configuration detail that matters to federal security and procurement teams evaluating voice AI.
The practical change claimed by ElevenLabs is a reusable body of security evidence for U.S. agencies. Its announcement frames Class A around most non-sensitive agency use cases, while the fetched evidence does not establish that every ElevenLabs configuration is covered or that a federal agency has approved a particular production deployment.
What the certification covers
The ElevenLabs announcement names a bounded set of services: the ElevenAgents platform and the company's Text to Speech and Speech to Text APIs. It also attaches two operating conditions to that scope: Zero Retention Mode and U.S. data residency.
That boundary is central to the story. The certification is not presented as a blanket statement about every ElevenLabs product, storage mode, hosting region or customer architecture. A team considering the services would still need to confirm that its planned deployment matches the certified configuration and its own agency requirements.
What Class A changes for agency reviews
According to ElevenLabs, FedRAMP reviewed its security package and confirmed that the company met the FedRAMP 20x Class A requirements. The company says the full evidence package is available through its trust center, so procurement and security reviews can begin with validated material instead of starting from an empty file.
The announcement describes Class A as providing adequate information for most non-sensitive agency use cases. It gives benefits, permits, tax and service requests as possible agent workflows, alongside hearing transcription and more accessible public communications using speech tools. Those are examples of potential uses, not evidence that agencies have deployed or approved each one.
Why the configuration boundary matters
Voice systems can combine an agent, transcription, generated speech, call infrastructure and data storage. The source only assigns certification coverage to the named ElevenLabs services in the stated retention and residency configuration. Security reviewers therefore need to distinguish the certified components from the rest of an application and verify how data moves across the complete system.
The page does not provide, in the material fetched for this article, a list of federal customers using the certified configuration, a workload-specific authorization decision or evidence about production outcomes. Certification can reduce duplicated evidence collection, but it does not by itself answer whether a specific use case is appropriate, accessible, accurate or operationally reliable.
Evidence quality and what to watch
This report relies on one official company announcement. It is authoritative for what ElevenLabs says it achieved and for the product scope the company describes, but the certification claim was not independently corroborated through a separately fetched federal record in this run. That is why internal confidence remains medium.
The next useful checks are the public federal record for the certification, the exact package scope and status, and procurement documents showing how an agency applies the certified services to a real workload. Those records would clarify the difference between reusable security evidence and an agency's own authorization to operate.
Status
Confirmed company announcement. ElevenLabs says the certification is complete, with coverage limited to the named services under Zero Retention Mode and U.S. data residency.
Sources
Update note: Last reviewed 2026-10-06. We will revise this post if a federal record or the stated certification scope changes.
Sources
Drafted with AI assistance from source briefs; reviewed for citation completeness and label accuracy.