Anthropic
anthropic.com
Automated, point-in-time. Not independently reviewed or approved by TRUSTYCYBER unless expressly stated.
Anthropic scores 63/100 (C): approve once the conditions are confirmed in writing. 3 conditions to confirm in writing, starting with: a no-training commitment was found: confirm its scope in writing. Independently confirmed: ISO/IEC 42001, SOC 2 and ISO/IEC 27001.
Scanned 28 Aug 2026 · Reassessed 5 Nov 2026 (methodology 0.13.1) · Public evidence · Point-in-time
Before you sign
What to ask for: Confirm in writing that the no-training commitment covers every model, plan and data type you will use, including any opt-in sharing or feedback.
Why it matters, and the evidence
The sources include a no-training statement ("Anthropic’s Commercial Terms state that Anthropic may not train models on Customer Content (Inputs and Outputs) from the Services.") but do not settle the question for every model, plan and data type in scope.
“Anthropic may not train models on Customer Content from Services.”
“We may use your Inputs and Outputs to train and improve Anthropic AI models, unless you opt out through your account settings.”
Neither document is the contract, so the report cannot say which applies.
“data from Claude users who have opted in to have their data used for training,”
“Even if you opt-out, we will use Inputs and Outputs for model improvement when: (i) your conversations are flagged for safety review to improve our ability to detect harmful content, enforce our policies, or advance AI safety research, or (ii) you've explicitly reported the materials to us (for example via our feedback mechanisms).”
What to ask for: Get retention periods in writing for inputs, outputs, logs and backups, including any provider that runs the model.
Why it matters, and the evidence
The sources describe retention ("Under the Commercial Terms a recipient destroys the other party’s Confidential Information (which includes Customer Content) promptly on request, except where retained to comply with law or held as copies in automated back-up systems.") but not a complete window for every data type.
“Recipient will destroy Discloser’s Confidential Information promptly upon request, except where retained to comply with law or copies in Recipient’s automated back-up systems, which will remain subject to these obligations of confidentiality while maintained.”
“delete all copies of Customer Data (including Customer Personal Data) processed by Anthropic or any Subprocessors, except to the extent (i) Applicable Data Protection Laws or other applicable legal or regulatory requirements requires storage of the Customer Data, (ii) retention of the Customer Data by Anthropic is necessary to resolve a dispute between the parties, or (iii) retention of the Customer Data is necessary to combat harmful use of the Services.”
“delete all copies of Customer Data (including Customer Personal Data) processed by Anthropic or any Subprocessors, except to the extent (i) Applicable Data Protection Laws or other applicable legal or regulatory requirements requires storage of the Customer Data, (ii) retention of the Customer Data by Anthropic is necessary to resolve a dispute between the parties, or (iii) retention of the Customer Data is necessary to combat harmful use of the Services.”
“Storage Limitation : The duration is the term of the Agreement.”
“You also are able to delete individual conversations , which will be removed immediately from your conversation history and automatically deleted from our back-end within 30 days.”
“You also are able to delete individual conversations , which will be removed immediately from your conversation history and automatically deleted from our back-end within 30 days.”
“Anthropic retains your personal data for as long as reasonably necessary for the purposes and criteria outlined in this Privacy Policy and explained further in our privacy center”
“Claude Fable 5 requires 30-day data retention and is not available under zero data retention.”
“Claude Fable 5 requires 30-day data retention and is not available under zero data retention.”
What to ask for: Ask the vendor to confirm in writing which applies, and that the contract says the same.
Why it matters, and the evidence
The Privacy Policy \ Anthropic and the Anthropic’s Transparency Hub \ Anthropic say different things about this. Neither is the contract, so this report cannot say which applies.
Australian Government checksclick a check for the detail, directions checked 1 Nov 2026
PSPF directionsNone apply
No vendor-specific PSPF direction names this vendor or product (checked 2026-11-01).
IRAP assessmentNot collected
Not found in the sources this scan collected, which is not evidence the vendor has none. There is no public register of IRAP assessments to check this against: ASD lists assessors, not the services they assessed. Ask for the assessment report or letter of completion and check its date, the classification level, the services in scope and the controls tested. An assessment is not a certification and does not mean every control is met.
Hosting Certification FrameworkHost certified
The sources name Amazon Web Services, Google Australia and Microsoft as hosting infrastructure for Anthropic. Amazon Web Services' Cloud Service offering (Australian Regions) is certified at Strategic level; Google Australia's Cloud Service offering (Google Cloud Platform, Australian Regions) is certified at Strategic level; Microsoft's Microsoft Azure Cloud (Australian Regions) is certified at Strategic level. That certification covers the hosting, not Anthropic itself: confirm in the contract that your data is held in an Australian region on the certified service. DTA list checked 4 Nov 2026. The HCF is under reform: since 3 November 2025 the DTA has paused new registrations and supplementary assessments; existing certifications are not affected.
“Cloud computing resources from Amazon Web Services, Google Cloud Platform and Microsoft Azure, supported by development frameworks including PyTorch, JAX, and Triton.”
“Cloud computing resources from Amazon Web Services, Google Cloud Platform and Microsoft Azure, supported by development frameworks including PyTorch, JAX, and Triton.”
“Cloud computing resources from Amazon Web Services, Google Cloud Platform and Microsoft Azure, supported by development frameworks including PyTorch, JAX, and Triton.”
Data held in AustraliaNot collected
No Australian data residency commitment was found in the sources this scan collected. Ask where inputs, outputs, logs and backups are processed and stored, and whether processing can be kept in Australia.
What your organisation still has to do (7)
Every level of government
- Treat AI outputs, including meeting transcripts and summaries, as public records: capture or dispose of them under your records authority (the Archives Act 1983 for Commonwealth agencies, your state's records legislation for state agencies and councils). Source ↗
Commonwealth agencies
- Assess foreign ownership, control or influence risk for this technology asset, as PSPF Direction 001-2024 requires. Source ↗
- Apply the DTA Policy for the responsible use of AI in government: record the use case, assess its impact before deployment, and keep your AI transparency statement current. Source ↗
- Assess the service against the ISM controls for the highest classification of data you intend to use with it (OFFICIAL, OFFICIAL: Sensitive or PROTECTED). Source ↗
State agencies and councils
- Assess the service under your state's privacy law, which governs where residents' personal information may go: for example the PPIP Act 1998 in New South Wales, the Privacy and Data Protection Act 2014 in Victoria or the Information Privacy Act 2009 in Queensland.
- New South Wales agencies: complete the NSW AI Assessment Framework self-assessment, which Circular DCS-2024-04 makes mandatory. Source ↗
- Victorian public sector: follow Administrative Guideline 2024/07 on the safe and responsible use of generative AI. Source ↗
What to ask the vendor
Buyer questionsanswered only from the public evidence scanned, click a row for the answer
!Will they train on your data?Ask the vendor
For consumer services, even after a user opts out, Anthropic uses Inputs and Outputs for model improvement when conversations are flagged for safety review or the user explicitly reports them.
Requires written confirmation, see Before you sign ↓
“Anthropic may not train models on Customer Content from Services.”
“We may use your Inputs and Outputs to train and improve Anthropic AI models, unless you opt out through your account settings.”
Neither document is the contract, so the report cannot say which applies.
“data from Claude users who have opted in to have their data used for training,”
“Even if you opt-out, we will use Inputs and Outputs for model improvement when: (i) your conversations are flagged for safety review to improve our ability to detect harmful content, enforce our policies, or advance AI safety research, or (ii) you've explicitly reported the materials to us (for example via our feedback mechanisms).”
!How long do they keep your data?Ask the vendor
Anthropic states that Claude Fable 5 requires 30-day data retention.
Requires written confirmation, see Before you sign ↓
“Recipient will destroy Discloser’s Confidential Information promptly upon request, except where retained to comply with law or copies in Recipient’s automated back-up systems, which will remain subject to these obligations of confidentiality while maintained.”
“delete all copies of Customer Data (including Customer Personal Data) processed by Anthropic or any Subprocessors, except to the extent (i) Applicable Data Protection Laws or other applicable legal or regulatory requirements requires storage of the Customer Data, (ii) retention of the Customer Data by Anthropic is necessary to resolve a dispute between the parties, or (iii) retention of the Customer Data is necessary to combat harmful use of the Services.”
“delete all copies of Customer Data (including Customer Personal Data) processed by Anthropic or any Subprocessors, except to the extent (i) Applicable Data Protection Laws or other applicable legal or regulatory requirements requires storage of the Customer Data, (ii) retention of the Customer Data by Anthropic is necessary to resolve a dispute between the parties, or (iii) retention of the Customer Data is necessary to combat harmful use of the Services.”
“Storage Limitation : The duration is the term of the Agreement.”
“You also are able to delete individual conversations , which will be removed immediately from your conversation history and automatically deleted from our back-end within 30 days.”
“You also are able to delete individual conversations , which will be removed immediately from your conversation history and automatically deleted from our back-end within 30 days.”
“Anthropic retains your personal data for as long as reasonably necessary for the purposes and criteria outlined in this Privacy Policy and explained further in our privacy center”
“Claude Fable 5 requires 30-day data retention and is not available under zero data retention.”
“Claude Fable 5 requires 30-day data retention and is not available under zero data retention.”
!Who else can access your data?Ask the vendor
Schedule 4 of the DPA points to Anthropic’s published list of subprocessors at anthropic.com/subprocessors; the list itself was not among the collected documents.
Confirm in writing: Ask the vendor to state this in writing before signing.
“B.3.a. “sell” or “share” Customer Personal Data, as defined by Applicable Data Protection Laws;”
“retain, use, or disclose Customer Personal Data outside of the direct business relationship and for any purpose other than for the business purposes specified in Part B of Schedule 1 or as otherwise permitted by Applicable Data Protection Laws;”
“Anthropic’s list of subprocessors is available at https://www.anthropic.com/subprocessors”
“You can find information about the third parties Anthropic engages to help us process personal data here”
“In the event that Anthropic wishes to appoint an additional Subprocessor: (a) Anthropic will provide Customer reasonable notice of the new Subprocessor prior to giving the Subprocessor access to Customer Personal Data;”
“Customer may, on the basis of reasonable data privacy or data security concerns, object to Anthropic’s use of such Subprocessor by providing Anthropic with written notice of the objection within fifteen (15) days of the date of such notice, or Customer is deemed to consent to the new Subprocessor.”
“Claude may send your Inputs, Outputs, and instructions to Third-Party Services to perform actions on your behalf (such as reading files, sending messages, or retrieving information).”
!Where is your data processed?Ask the vendor
For consumer services, personal data is transferred to Anthropic’s servers in the US or to other countries outside the EEA and the UK.
Confirm in writing: Ask the vendor to state this in writing before signing.
“Anthropic is a global company, and when you access our Services, your personal data is transferred to our servers in the US, or to other countries outside the European Economic Area (“EEA”) and the UK for the purposes outlined in Section 2 of this Privacy Policy, for example to provide you with the Service, or to train our models, or to conduct research.”
“We've introduced data residency controls , allowing you to specify where model inference runs with the inference_geo parameter.”
“US-only inference is available at 1.1x pricing for models released after February 1, 2026.”
!What happens in a security incident?Ask the vendor
The DPA commits Anthropic to notify the customer in writing without undue delay, and in any event within 48 hours, after becoming aware of a Security Breach.
Confirm in writing: Ask the vendor to state this in writing before signing.
“Anthropic will notify Customer in writing without undue delay, but in any event within 48 hours, after becoming aware of any Security Breach, and will assist Customer in complying with Customer’s obligations under Applicable Data Protection laws by reasonably cooperating with Customer’s investigation of the Security Breach.”
“These incident response procedures are tested annually.”
Email to send the vendor7 items to confirm in writing
Hello Anthropic team, We are assessing Anthropic as part of our supplier review. Before we proceed, please confirm the following in writing: 1. Do you use our data (inputs, outputs or uploaded files) to train or improve your models? Please confirm in writing, including any opt-out and whether it applies to every plan. 2. How long do you retain our data, in days, for each type (inputs, outputs, logs and backups), and how is it deleted? 3. Please address the following: The vendor's documents disagree: Consumer model training. 4. Please provide your current, dated subprocessor list and explain how you notify customers of changes. 5. In which countries and regions is our data processed and stored, and can processing be limited to a region we choose? 6. What is your commitment to notify customers of a security incident affecting our data, including the timeframe? 7. Which inference_geo values are available, is an Australian geography offered, and where is data stored at rest for API and Enterprise customers? A written reply to each point, or a link to where it is documented, is enough. Where a point is covered by a certification or independent report, please include the certificate number or the report and its date. Thank you,
How the score is built
Assurance dimensionsweighted components of one score, not eight separate ratings
Organisation & AI governance15% of the score77
Who owns AI risk: policies, responsible-AI principles, human oversight and administrative controls.
Assessed against ISO 42001 · NIST AI RMF · AIUC-1
“Version 3.4 and redline (effective July 8, 2026)”
“Version 3.2 of our RSP authorizes the LTBT to request external review of Risk Reports, gives the LTBT the power to approve our selection of external reviewers, and formalizes a requirement that we provide the LTBT with regular briefings.”
“Executive Risk Council: Establish an Executive Risk Council sponsored by executive leadership to oversee security programs.”
“This new version of the RSP involves the publication of Frontier Safety Roadmaps with detailed safety goals, and Risk Reports that quantify risk across all our deployed models.”
“0 1 Model Report”
“As we learn more about its operation in the real world, we expect to make further changes, all of which will be logged both on this page and in a changelog in the policy document itself.”
“We posted an update to our RSP Noncompliance Reporting and Anti-Retaliation Policy, which we released internally in February 2026.”
“Follow a risk-based approach aligned with ISO 27001 standard.”
“0 2 System Trust and Reporting”
“0 3 Voluntary Commitments”
Governance & accountability: the certifications above are independently corroborated, but they cover the management system — these specific disclosures are the vendor's own statements, not yet independently verified.
AI system15% of the score53
What AI the vendor actually runs and where it sits in the product, the map everything else is judged against.
Assessed against ISO 42001 · NIST AI RMF
“Claude is an artificial intelligence, trained by Anthropic using Constitutional AI to be safe, accurate, and secure — the trusted assistant for you to do your best work.”
“We used a technique called Constitutional AI to align Claude with human values during reinforcement learning.”
“Integrate any context or tool through connectors with remote MCP”
“Memory across conversations”
“Depending on the permissions you grant Claude, some Outputs may result in actions with effects outside the Services, such as sending communications, modifying files, or interacting with third-party services on your behalf.”
“It's available on the Claude API, Claude in Amazon Bedrock , Claude Platform on AWS , Claude on Google Cloud , and Claude in Microsoft Foundry”
The vendor’s own evaluation of its AI features before and after release, evals, red-teaming, monitoring. A model supplier’s testing of its own model counts only in part.
Assessed against NIST GenAI · NIST AI RMF · ISO 42001
“System cards document the capabilities, safety evaluations, and responsible deployment decisions for Claude models.”
“Additional evaluations were conducted as part of our safety process; for our complete publicly reported evaluation results, please refer to the full system card”
“The IPI benchmark was built in partnership with Gray Swan, the UK AI Security Institute, the US Center for AI Standards and Innovation, and other model developers.”
“Our Responsible Scaling Policy (RSP) evaluation process is designed to systematically assess our models' capabilities in areas where they could pose catastrophic risks before we release them.”
“We worked with several external testers to get additional evidence about the difficulty of breaking our cyber classifiers.”
“Partner with a diverse range of external red team and penetration testing experts.”
“Anthropic engages qualified external assessors for the completion of annual penetration testing of systems that process Customer Data to identify vulnerabilities and attack vectors that can be used to exploit those systems, and identified vulnerabilities are addressed as part of Anthropic’s vulnerability management program.”
“Simulate sophisticated attacks, including insider threat and software supply chain compromise scenarios, to identify vulnerabilities.”
“In order to stay up to date on newly discovered harmful patterns, jailbreaks, and obfuscation techniques, the classifiers will be regularly updated using data from other elements of our deployment safeguards including our asynchronous monitoring system and insights from our incident response protocol, bug bounty program, and internal and external red-teaming efforts.”
How changes to models and AI features, including a change of underlying model, are controlled and communicated: versioning, notice, rollback.
Assessed against ISO 27001 · ISO 42001
“The Claude Platform release notes list changes to the Claude API, the client SDKs, and the Claude Console, newest first.”
“We announced the deprecation of the Claude Sonnet 4 model ( claude-sonnet-4-20250514 ) and the Claude Opus 4 model ( claude-opus-4-20250514 ), with retirement on the Claude API scheduled for June 15, 2026.”
“In client applications (the web interface and the desktop and mobile apps), the request is automatically redirected to the most recent Claude Opus model and the user is notified which model handled their request;”
Testing & evaluation: vendor-evidenced, not yet independently corroborated.
Change management: vendor-evidenced, not yet independently corroborated.
Model10% of the score60
Which foundation models and model providers power the product. How model changes are disclosed is judged under change management.
Assessed against NIST GenAI · ISO 42001
“We've launched Claude Opus 5 ( claude-opus-5 ), a step-change improvement over Claude Opus 4.8.”
“Claude Fable 5 is built on Mythos 5, the underlying model behind this release.”
Model provider transparency: vendor-evidenced, not yet independently corroborated.
Customer data15% of the score40
Whether customer data trains models, how long it is retained, and where it is processed.
Assessed against ISO 27018 · OAIC APPs · EU AI Act
“Anthropic may not train models on Customer Content from Services.”
“delete all copies of Customer Data (including Customer Personal Data) processed by Anthropic or any Subprocessors, except to the extent (i) Applicable Data Protection Laws or other applicable legal or regulatory requirements requires storage of the Customer Data, (ii) retention of the Customer Data by Anthropic is necessary to resolve a dispute between the parties, or (iii) retention of the Customer Data is necessary to combat harmful use of the Services.”
“delete all copies of Customer Data (including Customer Personal Data) processed by Anthropic or any Subprocessors, except to the extent (i) Applicable Data Protection Laws or other applicable legal or regulatory requirements requires storage of the Customer Data, (ii) retention of the Customer Data by Anthropic is necessary to resolve a dispute between the parties, or (iii) retention of the Customer Data is necessary to combat harmful use of the Services.”
“Claude Fable 5 requires 30-day data retention and is not available under zero data retention.”
“We've introduced data residency controls , allowing you to specify where model inference runs with the inference_geo parameter.”
“US-only inference is available at 1.1x pricing for models released after February 1, 2026.”
“As between the parties and to the extent permitted by applicable law, Anthropic agrees that Customer (a) retains all rights to its Inputs, and (b) owns its Outputs.”
“As between the parties and to the extent permitted by applicable law, Anthropic agrees that Customer (a) retains all rights to its Inputs, and (b) owns its Outputs.”
“Anthropic disclaims any rights it receives to the Customer Content under these Terms.”
“Customer Content is Customer’s Confidential Information.”
“Recipient will destroy Discloser’s Confidential Information promptly upon request, except where retained to comply with law or copies in Recipient’s automated back-up systems, which will remain subject to these obligations of confidentiality while maintained.”
“Storage Limitation : The duration is the term of the Agreement.”
“B.3.a. “sell” or “share” Customer Personal Data, as defined by Applicable Data Protection Laws;”
“retain, use, or disclose Customer Personal Data outside of the direct business relationship and for any purpose other than for the business purposes specified in Part B of Schedule 1 or as otherwise permitted by Applicable Data Protection Laws;”
“This Privacy Policy does not apply to content that we process on behalf of customers of our business offerings, such as our Enterprise accounts.”
“We may use your Inputs and Outputs to train and improve Anthropic AI models, unless you opt out through your account settings.”
Neither document is the contract, so the report cannot say which applies.
“data from Claude users who have opted in to have their data used for training,”
“Even if you opt-out, we will use Inputs and Outputs for model improvement when: (i) your conversations are flagged for safety review to improve our ability to detect harmful content, enforce our policies, or advance AI safety research, or (ii) you've explicitly reported the materials to us (for example via our feedback mechanisms).”
“You also are able to delete individual conversations , which will be removed immediately from your conversation history and automatically deleted from our back-end within 30 days.”
“You also are able to delete individual conversations , which will be removed immediately from your conversation history and automatically deleted from our back-end within 30 days.”
“Anthropic retains your personal data for as long as reasonably necessary for the purposes and criteria outlined in this Privacy Policy and explained further in our privacy center”
“Anthropic is a global company, and when you access our Services, your personal data is transferred to our servers in the US, or to other countries outside the European Economic Area (“EEA”) and the UK for the purposes outlined in Section 2 of this Privacy Policy, for example to provide you with the Service, or to train our models, or to conduct research.”
“Claude may send your Inputs, Outputs, and instructions to Third-Party Services to perform actions on your behalf (such as reading files, sending messages, or retrieving information).”
“Claude Fable 5 requires 30-day data retention and is not available under zero data retention.”
AI supply chain10% of the score40
The named chain of AI and infrastructure providers behind the product, and how it is disclosed.
Assessed against ISO 27001 · ISO 27017
“Anthropic’s list of subprocessors is available at https://www.anthropic.com/subprocessors”
“You can find information about the third parties Anthropic engages to help us process personal data here”
“A.9. "Subprocessor" means an entity engaged by Anthropic to process Customer Personal Data.”
“Anthropic will: (a) enter into a contractual agreement with each Subprocessor imposing data protection obligations that are substantially as protective as Anthropic’s obligations under this DPA to the extent applicable to the nature of the services provided by Subprocessor;”
“In the event that Anthropic wishes to appoint an additional Subprocessor: (a) Anthropic will provide Customer reasonable notice of the new Subprocessor prior to giving the Subprocessor access to Customer Personal Data;”
“Customer may, on the basis of reasonable data privacy or data security concerns, object to Anthropic’s use of such Subprocessor by providing Anthropic with written notice of the objection within fifteen (15) days of the date of such notice, or Customer is deemed to consent to the new Subprocessor.”
“Cloud computing resources from Amazon Web Services, Google Cloud Platform and Microsoft Azure, supported by development frameworks including PyTorch, JAX, and Triton.”
“Cloud computing resources from Amazon Web Services, Google Cloud Platform and Microsoft Azure, supported by development frameworks including PyTorch, JAX, and Triton.”
“Cloud computing resources from Amazon Web Services, Google Cloud Platform and Microsoft Azure, supported by development frameworks including PyTorch, JAX, and Triton.”
“The new Claude in Amazon Bedrock endpoint at /anthropic/v1/messages uses the same request shape as the first-party Claude API and runs on AWS-managed infrastructure with zero operator access.”
Security foundation15% of the score85
How security problems are found, reported and handled: disclosure channels, bug bounty, incident response.
Assessed against ISO 27001 · NIST AI RMF
“Contact: https://hackerone.com/4f1f16ba-10d3-4d09-9ecc-c721aad90f24/embedded_submissions/new”
“Policy: https://www.anthropic.com/responsible-disclosure-policy”
“It does not generally cover reports to Anthropic by external researchers, which are governed by Anthropic’s Responsible Disclosure Policy.”
“Anthropic will notify Customer in writing without undue delay, but in any event within 48 hours, after becoming aware of any Security Breach, and will assist Customer in complying with Customer’s obligations under Applicable Data Protection laws by reasonably cooperating with Customer’s investigation of the Security Breach.”
“These incident response procedures are tested annually.”
“Conduct consistent scanning of all third-party dependencies and maintain a comprehensive vulnerability data repository.”
“Conduct consistent scanning of all third-party dependencies and maintain a comprehensive vulnerability data repository.”
“Proactively scan all incoming packages to enable download of only secure packages and reduce risk of introducing vulnerabilities.”
“In order to stay up to date on newly discovered harmful patterns, jailbreaks, and obfuscation techniques, the classifiers will be regularly updated using data from other elements of our deployment safeguards including our asynchronous monitoring system and insights from our incident response protocol, bug bounty program, and internal and external red-teaming efforts.”
“In order to stay up to date on newly discovered harmful patterns, jailbreaks, and obfuscation techniques, the classifiers will be regularly updated using data from other elements of our deployment safeguards including our asynchronous monitoring system and insights from our incident response protocol, bug bounty program, and internal and external red-teaming efforts.”
“These operating principles apply to vulnerabilities Anthropic discovers in open-source software and in closed-source software where Anthropic has obtained appropriate authorization to conduct security research.”
“Anthropic aims to follow the industry standard 90-day disclosure deadline, provide human-reviewed reports with suggested fixes where we can, and pace our submissions to what maintainers can actually absorb.”
Vulnerability & incident handling: the certifications above are independently corroborated, but they cover the management system — these specific disclosures are the vendor's own statements, not yet independently verified.
Independent assurance evidence10% of the score92
Whether claimed certifications and attestations are independently corroborated, checked against official registries and issuing bodies. Which certifications, and what each one does and does not establish, is stated per certificate.
Assessed against ISO 17021-1 · ISO 42006 · ISO 27006-1 · ISO 27008
“Anthropic is audited annually against known, established industry standards performed by external auditors.”
“Upon Customer’s written request, and subject to the confidentiality obligations set forth in the Agreement, Anthropic will provide Customer with such audit reports or certificates applicable to the Services ( e.g. , SOC 2 report), to the extent available, or such other information reasonably necessary to demonstrate compliance with this DPA.”
“Anthropic’s current certifications are available for Customer’s review at trust.anthropic.com”
Read from the registry record above, cited, not reproduced.
Read from the registry record above, cited, not reproduced.
Read from the registry record above, cited, not reproduced.
Document held and reviewed by TrustyCyber, cited, not reproduced.
Document held and reviewed by TrustyCyber, cited, not reproduced.
Independent assurance: vendor-evidenced, not yet independently corroborated.
Legal & contractual10% of the score60
The contractual backbone a buyer can rely on, terms, DPAs, privacy commitments and regulatory posture.
Assessed against EU AI Act · OAIC APPs
“Data submitted through the Services will be processed in accordance with the Anthropic Data Processing Addendum (“ DPA ”), which is incorporated into these Terms by reference.”
“This Data Processing Addendum ( “DPA” ) is incorporated into and forms part of the Anthropic Commercial Terms of Service”
“Customer and its Users may only use the Services in compliance with these Terms, including (a) the Usage Policy (“ Usage Policy ”, which was previously referred to as the Acceptable Use Policy), (b) our policy on the countries and regions Anthropic currently supports (“ Supported Regions Policy ”) and (c) our Service Specific Terms , each of which is incorporated by reference into these Terms.”
“The parties agree that, to the extent required by Applicable Data Protection Laws, the terms of the SCCs Module Two (controller to processor) and/or Module Three (processor to processor), as completed as described in Schedule 3 of this DPA, are hereby incorporated by reference and will be deemed to have been executed by the parties.”
“Services under these Terms are not for consumer use.”
“Customer and its Users may only use the Services in compliance with these Terms, including (a) the Usage Policy (“ Usage Policy ”, which was previously referred to as the Acceptable Use Policy), (b) our policy on the countries and regions Anthropic currently supports (“ Supported Regions Policy ”) and (c) our Service Specific Terms , each of which is incorporated by reference into these Terms.”
“Customer and its Users may only use the Services in compliance with these Terms, including (a) the Usage Policy (“ Usage Policy ”, which was previously referred to as the Acceptable Use Policy), (b) our policy on the countries and regions Anthropic currently supports (“ Supported Regions Policy ”) and (c) our Service Specific Terms , each of which is incorporated by reference into these Terms.”
“Anthropic may update these Terms at any time, to be effective 30 days after the updates are posted by Anthropic or Customer otherwise receives Notice, except that updates made in response to changes to law or regulation take effect immediately upon posting or Notice.”
Legal & contractual transparency: vendor-evidenced, not yet independently corroborated.
Not graded: Agent, not applicable to this scan.
What would strengthen assuranceeach figure is from today's score, resolving one changes the others
Resolving an evidence gap strengthens the dimension it belongs to, and with it the score: 63 → up to 80 with every identified gap resolved.
The evidence behind the result
Key findingsclick a row for the evidence
✓Contractual no-training and deletion commitments for commercial customersStrong
The Commercial Terms state that Anthropic may not train models on Customer Content, and the DPA commits Anthropic to delete Customer Data, including Subprocessor copies, within 30 days of termination, with narrow legal, dispute and harmful-use exceptions.
Training and retention commitments in the contract, not only on a marketing page, are enforceable and answer the two most common buyer data questions for API and other commercial use.
“Anthropic may not train models on Customer Content from Services.”
“delete all copies of Customer Data (including Customer Personal Data) processed by Anthropic or any Subprocessors, except to the extent (i) Applicable Data Protection Laws or other applicable legal or regulatory requirements requires storage of the Customer Data, (ii) retention of the Customer Data by Anthropic is necessary to resolve a dispute between the parties, or (iii) retention of the Customer Data is necessary to combat harmful use of the Services.”
“delete all copies of Customer Data (including Customer Personal Data) processed by Anthropic or any Subprocessors, except to the extent (i) Applicable Data Protection Laws or other applicable legal or regulatory requirements requires storage of the Customer Data, (ii) retention of the Customer Data by Anthropic is necessary to resolve a dispute between the parties, or (iii) retention of the Customer Data is necessary to combat harmful use of the Services.”
✓48-hour breach notification and a published disclosure channelStrong
The DPA commits Anthropic to notify customers in writing within 48 hours of becoming aware of a Security Breach, with incident procedures tested annually; security.txt routes vulnerability reports to HackerOne under a responsible disclosure policy.
A stated contractual notification window plus a public reporting channel meets the vulnerability and incident handling expectation in full.
“Anthropic will notify Customer in writing without undue delay, but in any event within 48 hours, after becoming aware of any Security Breach, and will assist Customer in complying with Customer’s obligations under Applicable Data Protection laws by reasonably cooperating with Customer’s investigation of the Security Breach.”
“These incident response procedures are tested annually.”
“Contact: https://hackerone.com/4f1f16ba-10d3-4d09-9ecc-c721aad90f24/embedded_submissions/new”
“Policy: https://www.anthropic.com/responsible-disclosure-policy”
“It does not generally cover reports to Anthropic by external researchers, which are governed by Anthropic’s Responsible Disclosure Policy.”
!Consumer training consent described as opt-out in one document and opt-in in anotherGap
The consumer Privacy Policy says Inputs and Outputs may be used for training unless the user opts out, and are used even after opt-out when flagged or reported. The Transparency Hub model reports describe training data as coming from Claude users who have opted in. Both statements concern consumer services, not the Commercial Terms Services.
Staff using consumer Claude accounts for work could expose organisational content to training unless the opt-out is set; the two descriptions of the consent model disagree.
Question for vendor: Is consumer Claude training consent opt-in or opt-out today, and which plans (Free, Pro, Max, Team, Enterprise) fall under the consumer Privacy Policy rather than the Commercial Terms?
“We may use your Inputs and Outputs to train and improve Anthropic AI models, unless you opt out through your account settings.”
Neither document is the contract, so the report cannot say which applies.
“data from Claude users who have opted in to have their data used for training,”
“Even if you opt-out, we will use Inputs and Outputs for model improvement when: (i) your conversations are flagged for safety review to improve our ability to detect harmful content, enforce our policies, or advance AI safety research, or (ii) you've explicitly reported the materials to us (for example via our feedback mechanisms).”
!Subprocessor register referenced but not reviewedGap
DPA Schedule 4 points to a published subprocessor list at anthropic.com/subprocessors, with prior notice and a 15-day objection right for new subprocessors, but the list itself was not among the collected documents, so the named providers in the data path could not be checked.
Without the list, a buyer cannot confirm which hosting and AI providers process their data or where.
Question for vendor: Please confirm the current subprocessor list and the processing locations of each subprocessor for API and Enterprise customers.
“Anthropic’s list of subprocessors is available at https://www.anthropic.com/subprocessors”
“You can find information about the third parties Anthropic engages to help us process personal data here”
“In the event that Anthropic wishes to appoint an additional Subprocessor: (a) Anthropic will provide Customer reasonable notice of the new Subprocessor prior to giving the Subprocessor access to Customer Personal Data;”
“Anthropic will: (a) enter into a contractual agreement with each Subprocessor imposing data protection obligations that are substantially as protective as Anthropic’s obligations under this DPA to the extent applicable to the nature of the services provided by Subprocessor;”
!Inference location control names only a US option; Fable 5 needs 30-day retentionGap
The release notes describe an inference_geo data residency control and US-only inference at 1.1x pricing; no Australian option is named in the collected documents. Claude Fable 5 requires 30-day data retention and is unavailable under zero data retention.
Australian buyers with residency or zero-retention requirements need to confirm which geographies are available and avoid models that require retention.
Question for vendor: Which inference_geo values are available, is an Australian geography offered, and where is data stored at rest for API and Enterprise customers?
“We've introduced data residency controls , allowing you to specify where model inference runs with the inference_geo parameter.”
“US-only inference is available at 1.1x pricing for models released after February 1, 2026.”
“Claude Fable 5 requires 30-day data retention and is not available under zero data retention.”
“Claude Fable 5 requires 30-day data retention and is not available under zero data retention.”
Assurance evidence: certifications 5, registry checks and verification ladders, click to view
Checked against Evidence held by TrustyCyber (not reproduced), 25 Aug 2026: Verified on the registry
The DPA offers the SOC 2 report on written request; the report itself is not among the collected public documents.
Checked against Evidence held by TrustyCyber (not reproduced), 25 Aug 2026: Verified on the registry
Checked against Evidence held by TrustyCyber (not reproduced), 25 Aug 2026: Verified on the registry
Checked against CSA STAR Registry, 28 Aug 2026: Verified on the registry
Checked against Evidence held by TrustyCyber (not reproduced), 5 Nov 2026: Verified on the registry
AI supply chainclick a node to focus it · drag to pan · zoom with the controls
View as list
What you inheritTrustyCyber's AI Shared Responsibility Matrix v1.0
This scan names the vendor’s AI supply chain without assessing the responsibility split, that depends on how you consume the product. Under the common managed models, a deployer inherits 16 of the matrix’s 41 responsibilities from providers, each with named evidence owed.
Map your inherited responsibilitiesSources and regulation
Sources 17, click to view
Appendix: standards landscape 8, click to view
Where each standard behind this assessment stands today, from TrustyCyber’s continuously maintained standards radar.
Want to go further?
This scan assesses Anthropic at an organisational level. Assurance can vary significantly between products, models and agents, enter the product you’re evaluating to run a deeper AI System Assurance Scan.
Scan a specific product →This automated result is based on publicly available information at the time of scanning. It is not an audit, certification, legal opinion or assurance engagement. Publicly unavailable evidence may materially change the conclusion. Some certification facts were confirmed by TrustyCyber from documents it holds; only facts that may be disclosed are stated, and no confidential content is reproduced.