Anthropic has announced Enterprise Frontier Safeguards (EFS), a new solution that combines the privacy of zero data retention with state-of-the-art safeguards for detecting misuse. The system stores monitoring data in cloud infrastructure controlled by the customer rather than by Anthropic. EFS will roll out in phases later this fall, and eligible customers will receive zero data retention on Fable 5 and Fable 5.1 until it is ready.
Addressing Frontier Model Security
Anthropic developed EFS with more than 100 customers across financial services, healthcare, manufacturing, telecom, law, retail, and the public sector. The company also worked with cloud partners Amazon Web Services, Google Cloud, and Microsoft Azure. EFS will be supported on Claude Code, Claude Enterprise, the Claude Platform, Amazon Bedrock, Google Agent Platform, and Microsoft Foundry.
The announcement addresses a security dilemma posed by highly capable models such as Claude Fable 5.1. Anthropic said it has seen substantial evidence of attempted misuse, ranging from fraud to sophisticated cyberattacks that can involve agents acting destructively. Because serious misuse can span multiple sessions and accounts, the company introduced 30-day data retention with Fable 5, while stating that it has never trained on enterprise data without explicit permission.
Designed with Regulated Industries
Many enterprises understood the safety value of data retention but found it difficult to adopt in regulated environments. Anthropic worked with security, product, compliance, and delivery teams to design a solution that provides both privacy and safety. The Analysis and Resilience Center for Systemic Risk, whose members include chief information security officers from major US banks, contributed to the process.
The company also gathered feedback from leaders at Comcast, KPMG, Mastercard, Salesforce, and Visa. These consultations spanned a quarter of the Fortune 100, every US global systemically important bank, and virtually every regulated industry. Customers asked for monitoring signals sent directly to their teams and for data stored under their own control.
How Enterprise Frontier Safeguards Works
With EFS, customers control how monitoring data is reviewed and stored across direct Claude access and major cloud partner platforms. Activity data used for monitoring can reside in the customer's own cloud account, such as Amazon S3, Azure Blob Storage, or Google Cloud Storage. Automated safety monitoring flags patterns that need attention, and those signals are sent directly to customers without any human review by Anthropic employees.
The automated systems analyze a rolling window of traffic for signs of serious misuse, including attempts to develop offensive cyber or biological capabilities and signs of stolen or leaked credentials. Customer-owned storage, customer-managed encryption keys, and fully automated review are each opt-in controls. Anthropic said none of these options change model behavior, API pricing, or rate limits.
Pricing and Availability
Anthropic does not charge for Enterprise Frontier Safeguards. If customers choose to store data in their own cloud account, their cloud provider bills them for storage, reads, writes, and data egress fees in the same way as any other resource. The company aims to make EFS broadly available later this fall and is accepting access requests through an online form.
Enterprise Frontier Safeguards represents an effort to close the gap between enterprise privacy requirements and frontier model safety monitoring. By allowing customer-controlled storage and automated, customer-directed review, Anthropic aims to serve regulated industries without weakening its safeguards. The phased rollout will test how this model performs across major cloud platforms and enterprise environments.