Start using deepl now for translations your customers can trust; SOC 2 Type II confirms robust, ongoing controls over data handling. visitantes access the plataforma with MFA and encryption in transit, and clique to grant access only for approved perfis and categorias of data. If you need changes, comunicar-nos and we respond imediatamente.
The SOC 2 Type II report covers five categories of controls–security, confidentiality, privacy, processing integrity, and availability–and shows how sensíveis data is protected, how perfis are managed, and how contribuições are auditable on the plataforma.
For directors and security teams, the report informs risk decisions by your diretor of security. It outlines how controls devem be implemented, including vendor access, incident response, and data retention, and provides a practical path to align translation workflows with compliance needs. Your team podem ensure tratar data responsibly while keeping sensíveis information under the control of titular data owners, with access defined by perfis and categorias.
To start clique the security overview to download the Type II report, or comunicar-nos to schedule a tailored rollout for your organisation. If you manage a global site, segment content by categorias and assign contribuições to teams with explicit titular rights and perfis aligned to job roles; visitantes from external partners can be granted restricted access and you can monitor sensíveis data in real time imediatamente.
Scope of SOC 2 Type II Certification for DeepL's Translation Platform
Define and implement the formal scope of SOC 2 Type II for DeepL's Translation Platform and maintain continuous monitoring across data flows, system components, and governance processes.
We poderemos demonstrate alignment by mapping controls to the five trust principles and documenting the relação between the platform and data subjects, including titulares and indivíduos, with attention to conhecidos data and conteúdos públicos (públicas). Claro, this approach supports auditability and transparent reporting for customers and auditors, and aligns with the abril cycle for ongoing readiness.
Scope Boundaries
- Data processed on the DeepL Translation Platform includes inputs, translations, outputs, and metadata (estes data), with sensíveis data guarded by criptografia in transit and at rest.
- Data categories span inputs and outputs from públicas and conhecidas sources, related to titulares and indivíduos, with a clear relação between user data and translations.
- Storage and processing occur across multiple regiões and environments; armazenamento and atualização are controlled, with abril-driven reviews to ensure alignment with policy changes.
- Access control and revocation rules ensure that only authorized personnel access data; the controladora landis defines roles, responsibilities, and revogação procedures.
- Auditability is built with detailed auditoria trails and retention policies to support the period under review.
Controls and Operational Details
- Governance and policy alignment: map SOC 2 criteria to the five principles; policies deve be documented and exigido to provide evidence during audits.
- Encryption and key management: criptografia in transit and at rest; cryptographic keys are rotated regularly and maintained by a dedicated team; manutenção of keys and rotation schedule is enforced.
- Change management and maintenance: changes undergo testing, approval, and documented rollbacks; abril cycles are used to verify updates in production without disrupting service.
- Access management and revocation: implement least-privilege access, enforce revogação of access when roles change or employment ends; the controladora landis ensures policy adherence.
- Monitoring, auditing, and evidence collection: centralized logs, regular reviews, and anomaly detection; the auditoria trail is retained for the designated period; envie artifacts through approved secure channels when requested.
- Data subject rights and privacy: support for titulares and indivíduos to exercise rights; processes document consents, data minimization, and handling of conhecidos data, including racial data where applicable.
- Vendor and data-sharing controls: contractually govern third-party processors; minimize exposure to publicly known datasets (públicas) and conhecidos sources, with ongoing monitoring and documentation.
Specific Data Collected During Translation: Inputs, Metadata, and Usage Details
Recommendation: At the momento of submission, utiliza strict validation to corrigir inputs before processing, reducing exposure and protecting privadas data, while obrigaçao to obtain consent remains explicit.
Inputs include user-provided text, documents (PDF, Word, and other formats), glossaries, and supporting files. These items are transmitted eletrônico and received through a secure recepção path, with verificações in place to support accuracy and auditability. When users preenche forms or provide context to improve precision, the system collects only what is necessary to deliver the service and to minimize risk.
Metadata covers language pair, timestamps, source and target language codes, device type, app version, locale, and reception context. Verificações run on arrival to confirm data integrity, and controls ensure data handling stays within the following boundaries, including limits on usage and access tied to the specific translation task. The seguinte data categories help with troubleshooting and accountability without exposing broader datasets.
Usos include translation delivery, quality checks, and model improvements. Data are used for verificações and controlled trainings, with escolhas that restrict sharing to authorized partes and, quando applicable, outras comunicações with explicit consent. We solicitamos consent for uses beyond service delivery; privadas data remain protected, can be preenche to reflect user preferences, and permaneçam encrypted to defend vida and mitigate riscos.
Data Processing Purposes: How Your Information Is Used in Translation
Data Processing Purposes drive how we handle information during translation. We use content to deliver translations, to improve terminology and quality, to monitor performance, and to provide atualizações and notícias to consumers who opt in. We uphold a claro policy for consent and limit usos to the minimum needed for service.
We collect: content you provide for translation, language pairs, timestamps, and usage data such as números and interaction metrics. We also capture device information and anonymized identifiers to help detect abuse and improve reliability. Content that falls into religiosa categories is handled with extra care and only when necessary for delivery or compliance.
Usos include delivering translations, refining glossaries, improving accuracy, and aligning terminology across language pairs. Data is utilized to improve models and workflows, and to determine routing across canais. When needed, we send atualizações to customers and issue notícias about policy changes or service levels.
Legal and sharing: Data may be accessed by empresas and trusted service providers to operate the service, process pagamentos, and maintain security. If authorities request data, we comply with due process and ensure authorization (autorização) is documented. Data that is vinculado to your account is protected by access controls and kept only as long as required by law and our retention standards.
You have control: You can select an opção to limit certain usos, request acesso to your data, and determine what data stays vinculado to your account. You may revoke authorization for specific processing at any time via your account settings. We provide transparent notices via notícias and atualizações when policies or practices change, and we describe data flows in clear terms to help you saber how your information is used.
Security and certification: We certificamos SOC 2 Type II and maintain robust controls across encryption, access management, monitoring, and incident response. Our canal support team can answer questions about how data is used in translation and how you can exercise control over your information through the available options.
Data Retention, Deletion, and Anonymization Practices
Set a 12-month default retention for all non-PII data and auto-delete input, output, and logs after that window, with a formal legal-hold process for exceptions. This policy minimizes exposure while meeting SOC 2 Type II expectations and keeps operations efficient.
- Data inventory and labeling – Map data types: input text, translations, metadata, logs, and backups. Tag each with retention class: short-term (30–90 days), mid-term (180 days), long-term (12–24 months). Align with data minimization (coleta) and assign owners to ensure accountability.
- Deletion workflows – When a project ends or a user account is removed, trigger automated deletion of associated input, outputs, and analytics within 24–72 hours, except where a formal legal hold applies. Utilize cryptographic erasure in backups to prevent recovery.
- Anonymization and minimization – Before storing analytics or model improvements, apply pseudonymization and tokenization; strip or mask PII; store only aggregated statistics. IP addresses should be anonymized by truncating the last octet; use field-level masking for names and emails. This applies data protection as a core capability (aplicam).
- Backups and disaster recovery – Backups inherit the same retention windows; implement purge sweeps and periodic verification to ensure deleted data cannot be restored beyond retention timelines. Distribuído across regions with strict access controls to limit exposure.
- Data access and accountability – Enforce least-privilege access; maintain an audit trail (auditoria) of access, deletion, and anonymization actions; enable alerts for deletion requests or policy violations. Provide an informativo that summarizes policy status.
- User-facing controls – Offer a self-service interface to request data export or deletion (aviso de exclusão) and to customize retention preferences. Utilize opt-out options for training data and display deletion milestones in the informativo panel.
contas and canal data streams follow the same retention controls; fornece clear guidance on propriedade and responsible group ownership. When entendido about a request to interrompa data collection, we respond quickly and document the outcome for the grupo. Dependem policies on coleta and análises, and rastrear data lineage across environments remains a core practice (infelizmente in some edge cases). If lawful requests arise from policiaiais authorities or internacionais bodies, we respond with minimal disclosure and full audit trails (auditoria). Our approach keeps data protected, and chúng tôi continuously monitor for deviations to maintain compliance (somos cumprindo) and keep customers informed via the informativo.
Access Control and Internal Handling: Who Can Access Your Data
Limit access to data on a need-to-know basis and enforce multi-factor authentication for all sessions, including transmissão paths used by deepl and parceiro networks. Define roles with a clear função: admin, security, translator, reviewer, and suporte, and ensure access is granted only for the task at hand and revoked automatically when the task ends, claro and auditable. This approach oferece visibility into who accessed what and provides traceability for every action.
Access follows a documented processo of provisioning and deprovisioning, with role-based permissions and token-based sessions that expire on a short prazo (typically 90 days). Usamos senha management and automatico MFA where possible, and we use logs to identificar suspicious activity and tomar action quickly.
Internal handling keeps data within controlled segments; transmissão happens through encrypted channels and is not exposed to público networks. We implement ddos protections, monitor for anomálias, and tomar immediate action to contain any incident. This approach keeps atividades claras, with comentarios from the team and preocupações addressed within prazo.
We maintain auditable trails of access decisions and run algumas internal reviews to ensure consistency across teams. Partir desses padrões, we escalate through the formal processo and respond within o prazo definido, maintaining nacional alignment and a culture of transparente handling.
We coordinate with parceiros such as deepl and monitor security using Reuters-style incident reporting to keep customers informed. We operate with defined medidas to prevent ddos and data leakage, and tomar steps to isolate affected components and destes learnings to policy updates. If needed, we report to national authorities and continue closure with feedback to the client.
Security Measures: Encryption, Network Protections, and Audit Trails
Implement AES-256 encryption for data at rest (armazenamento) and TLS 1.3 for data in transit, and utilizar envelope encryption with a centralized KMS. Use per-tenant keys and enforce automatic rotation every 90 days to minimize risk. Capture origem and lineage with clear provenance, and attach inseridas metadata alongside content to prevent leakage. Require confirmação of identities before any decryption and enforce least-privilege access. In casos de incidentes, isolate affected datasets and trigger the incident-response runbook. All encryption activities feed concise entries in reports and dashboards on the painel, summarized by cards to aid quick reviews.
Design network protections with zero-trust principles: private endpoints, VPC segmentation, next-gen firewall, WAF, DDoS protection, and mutual-TLS for service-to-service calls. Limit data movement with strict ingress/egress rules and monitor with artificial intelligence-powered detection. Policiais compliance reviews ensure access controls stay aligned with policy; coletas telemetry from all network nodes feed a centralized SIEM. Difusão of sensitive data is prevented by data loss prevention rules; aqui and páginas document the enforcement, while the painel uses cards to show real-time status. Campanhas de awareness educate teams, and associações with vendors are listed in the policy. Logs disponibilizado to responders as part of the security workflow.
Audit trails are tamper-evident and stored in an immutable log store. The controls descritos require cryptographic signing and append-only storage; foram written in real time to a centralized SIEM and disponibilizado to the security team via the painel. Logs coletam details such as user, role, origem, timestamp, and source IP, including decryption events for full traceability. Páginas summarize incidents, with cards (cards) highlighting importantes metrics. Supplemental (supplemental) controls ensure retention and compliance; infelizmente, some data may be restricted by policy in certain casoss, so campanhas help keep staff trained. Here (aqui) we provide incident-case reviews to assist investigators and associações with regulators.
| Aspect | Key Controls | Notes |
|---|---|---|
| Encryption | AES-256 at rest (armazenamento); TLS 1.3 in transit; utilizar envelope encryption; centralized KMS; per-tenant keys; key rotation every 90 days; origem tagging; inseridas metadata protection. |
foram implemented; importantes controls descritos; incluidas políticas de confirmação de acesso. |
| Network Protections | Zero-trust architecture; private endpoints; VPC segmentation; WAF; DDoS protection; mutual-TLS; ingress/egress controls; policiáis reviews; coletas telemetry. |
difusão de dados é restrita; aqui estão as páginas e painéis que mostram status em tempo real; cards exibem métricas importantes. |
| Audit Trails | Tamper-evident, append-only logs; cryptographic signing; logs foram centralizados em SIEM; origens, usuários, timestamps e ações coletados; disponibilizado via painel; páginas e cards destacam incidências e métricas. |
as descritos controles garantem rastreabilidade; associações com provedores estão registradas; suplementos de retenção são aplicados; infelizmente, algumas jurisdições impõem limites de retenção. |
Privacy Rights and How to Exercise Data Access, Correction, or Deletion
Start by logging into your perfil in our privacy center to access the data you fornecidas and to request corrections or deletion. The portal shows data by category (perfil, atividades) and provides links to view or export a structured copy. You can review processing steps (tarefas) and data lineage, ensuring you understand how information is used in translation workflows and quality checks. We determine the scope of each request based on your identifiers and the respective records, and we respect disposições that govern data across estados. Our team trabalha constantemente to protect privacy, maintain audit trails, and support respectful handling of your data. Access requires a unique authentication token (única) provided through secure channels. In emergências, contact the privacy desk through the secure channels available in the portal. Private data privadas are protected with strict controls, and we limit access to purpose-bound processing, with clear status updates and timeframes. We are committed to performance and qualidade across services while keeping your rights at the forefront, and we provide transparent explanations of what we retain and why. If you need assistance, the privacy team pode assist you.
We designed the process to be clear and actionable, so you can see exactly what data exists, where it came from, and how it’s used. The toolset, including links and cards that group data by category, helps you review efficiently while maintaining segurança and respeitando suas escolhas. We continuously monitor the workflow to ensure accuracy and timeliness, responding to changes in your profile or activities (atividades) with minimal disruption and maximum clarity.
How to access your data
To exercise the right to access your data, go to the privacy center and select Data Access. Verify your identity using the required methods; we may ask for additional details if needed. Review categories such as perfil, atividades, and translation logs; use the links to view or download a machine-readable export. The export will show data origin and processing details to help you understand what is held for you. If something looks missing or incorrect, indicará the corrections in the form and we will review promptly, following disposições that determine how data is collected and used across estados. You can also inspect the data privacy cards to better scan categories and confirm that the data is complete.
How to request corrections or deletion
To correct inaccuracies or request deletion, submit precise instructions in the same privacy center. Our team will verify your identity and apply corrections to the applicable records in your perfil and related logs. For deletion, specify the scope: which datasets, which products, and the time frame; we will assess and perform deletion where permitted by disposições, retention policies, and applicable laws in each estado. If full deletion conflicts with legal obligations, we will offer anonymization or restricted processing instead. You can cadastrar the request and track progress via links in the portal; we will provide updates seguindo cada etapa and respond within the legally required timeframe. We protect private data with robust security controls and ensure that handling respects your rights, maintaining performance and qualidade across services through these actions, using a secure ferramenta to manage the requests.
