Privacy Policy
Last updated: September 16, 2026
This Privacy Policy explains how NullRun ("NullRun", "we", "us", or "our") collects, uses, discloses, and otherwise processes personal data in connection with the NullRun website, control center, and enforcement-gateway services (collectively, the "Service"). It is read together with our Terms of Service and our Data Processing Addendum; in case of conflict on a data-protection matter, the DPA prevails.
The Service is offered only to persons in eligible jurisdictions (see Terms of Service §16). We do not knowingly accept signups from, or direct the Service at, data subjects in the European Economic Area, the United Kingdom, Switzerland, the People's Republic of China, the Republic of India, the Russian Federation, Iran, the Democratic People's Republic of Korea, Syria, Cuba, Belarus, or any other comprehensively-sanctioned jurisdiction. The United States and Canada were added to the eligible list on 2026-09-16 (see §16); CCPA / CPRA thresholds do not apply at the current pre-revenue stage. The Fortress geo-block at the application edge rejects requests from ineligible IP ranges before they touch the application server; where the name/email screening layer has been enabled for the deployment, account-name and account-email screening against the published sanctions lists (OFAC, EU, UK, UN consolidated) is applied on top of the geo-block.
Where the personal data protection law of a customer's jurisdiction applies (see §16 for the canonical list), we honour the rights and obligations set out there. Nothing in this Policy is intended to override a non-derogable right that the customer enjoys under the law of their habitual residence or place of business.
1. Roles and Scope
NullRun acts as a data processor on behalf of customers in respect of "Customer Data" (as defined in our Terms) and as an independent data controllerfor "Account Data", "Billing Data", and "Operational Data". Our role is determined per processing activity, not per customer relationship, so the same piece of information may be processed in different capacities depending on why we hold it. The table below maps these roles to the categories described in §2; in case of conflict between the table and the narrative, the table governs.
| Data category | Role | Primary basis | Primary purpose |
|---|---|---|---|
| Account Data (login, profile, 2FA) | Controller | Performance of contract | Authenticate you and run your account |
| Billing Data (subscription, tax IDs, invoices) | Controller | Performance of contract; legal obligation | Process payment and meet tax / accounting law |
| Customer Data (decisions, configs, prompts you choose to store) | Processor | Performance of contract (acting on customer instructions) | Deliver the Service per the customer’s documented instructions |
| Operational Data (access logs, auth events, rate-limit counters) | Controller | Legitimate interests | Secure the Service, prevent fraud, preserve integrity |
| Marketing Data (optional product emails) | Controller | Consent (opt-in) | Send product updates you have opted in to |
We do not collect, store, or proxy end-user Large Language Model (LLM) API keys. NullRun never receives, persists, or relays the API credentials you use to call upstream model providers. The Service enforces policy on decisions that you have already submitted through your own client; it does not sit on your identity path to those providers.
Where NullRun processes Customer Data as a processor, the parties' data-protection obligations (including instructions, sub-processor controls, security measures, breach notification, audit, and return or deletion on termination) are set out in our Data Processing Addendum (DPA), which is incorporated into the Terms by reference once accepted. See §16 below.
2. Information We Collect
We collect the categories of personal data described below.
2.1 Account data (controller)
- Email address (used as the primary login identifier).
- Display name (optional; what you enter in your profile).
- Password hash (Argon2id) — we never store plaintext passwords.
- OAuth identity metadata when you sign in with a third-party provider (GitHub, Google): provider name, provider user ID, and the email address the provider returns. We do not receive your password or refresh tokens for these providers.
- Two-factor authentication status and method (TOTP). We store only the shared secret encrypted at rest; we never see your one-time codes.
2.2 Organisation and billing data (controller)
- Organisation name, slug, and the user IDs of its members and invites.
- Plan tier (Lite, Starter, Growth, Scale) and subscription status.
- Billing data processed by our payment processor Polar.sh (see §6). NullRun does not store full payment-card numbers or CVVs. Polar returns to us a customer-ID, subscription-ID, billing email, and tax identifiers (where required by law) only.
- Invoices and receipts that Polar generates and that we surface inside the control center.
2.3 Service data (processor)
- API keys you generate inside the control center. The API-key secret value is shown to you once and stored hashed (Argon2id, with a per-key random salt) for verification; we cannot recover the plaintext after creation. A truncated SHA-256 fingerprint of the key is stored next to the hash so the key can be looked up in constant time; it is a lookup index, not a decryptable copy.
- The HMAC signing secret issued alongside each API key. It is a second credential, distinct from the key itself, and we keep it in readable form because it is what lets us verify a signature you send. It grants no access on its own and is never returned by the API after the one-time creation response.
- Workflow, policy, and detector configuration: names, JSON bodies, version numbers, and the user who created or last modified each artefact.
- Decision records: policy snapshot, detector output, decision verdict (allow / flag / deny), token-count bucket, and the agent / API-key / workflow identifiers associated with the call. The prompt payload or tool inputs your application submitted are not part of the decision record. Where your agent emits a trace, the trace may carry the request and response body; traces are a separate store and are covered by the retention window in §5.
- Alert-channel configuration, including Slack OAuth tokens (encrypted at rest with AES-256 via pgcrypto) and outbound webhook URLs.
- Audit log entries: who did what, when, from which IP, and the resulting state change. The audit log is append-only and hash-chained.
2.4 Operational and security data (controller)
- Server access logs: source IP, request method, path, response code, request duration, user agent, and request ID.
- Authentication events: login, logout, session refresh, password reset, and two-factor enrolment events. We record the fact of the event and a hashed representation of the email; the plaintext email is not retained in log storage.
- Rate-limit and abuse-prevention counters.
- Off-site backups of the production database, retained as described in §5.
2.5 What we do not collect
- End-user LLM API keys or model-provider credentials.
- Cross-site tracking identifiers, advertising IDs, or third-party marketing cookies. See our Cookie Policy.
- Biometric data (other than the optional TOTP secret used for second-factor authentication, which is encrypted at rest and discarded when you remove 2FA).
- Sensitive personal data under the laws referenced in §16 (race, religion, health, sexual orientation, genetic / biometric data, etc.). NullRun is not designed to process this data. Where Customer Data incidentally contains such material (for example, a prompt your application submits to the Service), you remain responsible for ensuring that you have a valid legal basis for that processing, and we will assist you with data-subject requests, deletion, and export in line with our DPA.
2.6 Prompt data minimisation
Prompts, completions, and tool inputs submitted to the Service can in principle contain personal data or other sensitive material. We apply a default of minimum-necessary capture:
- Decision records do not include the prompt payload, completion, or tool input — the record holds the policy snapshot, the detector output, the verdict, the token-count bucket, and the identifiers needed to correlate the call with your account. There is no setting that changes this.
- Where your agent emits a trace, the trace can carry the request body. Traces live in a separate store from the decision log, are retained for your plan's history window (§5), and are never written back into the decision record.
3. Legal Bases for Processing
The legal bases we rely on vary by jurisdiction and by data category. The canonical mapping for the categories described in §2 is the table in §1 above. The bases are:
- Performance of a contract: to provide the Service, authenticate you, process billing, and send you transactional emails (account verification, password reset, billing receipts, security alerts).
- Legitimate interests: to secure the Service, prevent abuse and fraud, debug issues, and improve platform reliability. We balance these interests against your rights and freedoms and provide opt-out mechanisms where reasonably possible.
- Compliance with a legal obligation: to comply with tax, accounting, anti-money-laundering, and law-enforcement obligations, and to respond to binding legal process.
- Consent (opt-in): for any optional processing for which we ask your consent (for example, optional product-marketing emails). You can withdraw consent at any time without affecting the lawfulness of processing carried out before withdrawal.
4. How We Use Information
We use personal data only for the following purposes:
- Operating, maintaining, securing, and improving the Service.
- Authenticating users and authorising actions inside the control center.
- Processing subscription payments and preventing billing fraud.
- Sending you service-critical notifications (security alerts, billing events, product-breaking changes). These are not optional while your account is active.
- Providing customer support and responding to your inquiries.
- Detecting, preventing, and responding to security incidents, abuse, and violations of our Terms.
- Complying with our legal obligations and enforcing our Terms.
Operational logs. Server access logs, authentication events, and rate-limit counters (§2.4) are processed on the basis of our legitimate interests. We have determined that this processing is necessary for security, fraud prevention, abuse detection, and the integrity of the Service, that the data is minimised to what those purposes strictly require (see §2.4), and that a reasonable person would expect this processing given the nature of the Service. Where you have an account, we balance these interests against your rights and freedoms and provide the retention controls in §5.
Prompt data. Your decision records never carry the prompt payload, completion, or tool input. Where your agent emits a trace, that trace may contain the request and response body; it is processed only to (a) let you review the call in the control center and (b) honour your deletion, export, and correction instructions. We do not use prompt bodies to train, fine-tune, evaluate, or otherwise improve any AI / ML model. See §2.6 for the capture posture and §5 for the retention window.
We do not use personal data for automated decision-making that produces legal or similarly significant effects concerning you. The allow / flag / deny decisions the Service records about your agents' actions are decisions made by your configured policy, not automated decisions made by NullRun about you as a person.
5. Retention
We retain personal data only for as long as necessary for the purposes for which we collected it, including for the purposes of satisfying any legal, accounting, or reporting requirements. Deletion requests are honoured within the windows below, subject to legal retention obligations and the standard backup-overwrite cycle.
- Account and organisation data: for the life of the account. On account deletion we delete controller data within 60 days, except where retention is required for legal, tax, security, or fraud-prevention reasons (for example, tax-invoice retention, anti-money-laundering records, or evidence needed to defend a pending legal claim). Where we retain data for one of these reasons, we isolate it from active processing and delete it once the underlying obligation or claim is resolved.
- Decision records and audit log: per the retention window of your plan (Lite: 3 days, Starter: 7 days, Growth: 30 days, Scale: 90 days, Enterprise: unlimited). On plan downgrade we retain prior records to the lower tier's window; on account deletion we delete all decision and audit records within 60 days, subject to the backup clause below. The canonical source for plan-level retention is the
planstable in our production database; where this Policy and the plan catalog disagree, the plan catalog governs, except where a different term is required by applicable law. - Server access and authentication logs: up to 14 days, and up to 7 days in the central log store, then deleted. We retain these logs only for the period strictly necessary to investigate security incidents, abuse, and operational integrity.
- Database backups: up to 7 days of rolling backups, after which they are overwritten. Where a deletion request lands within the 7-day backup window, the active record is deleted within the window in §5 above and the residual backup copy is overwritten on the normal cycle; we do not restore from backup to recover deleted user data on request.
- Billing records: for the period required by applicable tax law (typically 7 years in many jurisdictions, varying elsewhere). These records are not deleted on account closure where the law requires us to keep them; we isolate them from active processing and limit access to the finance function.
Nothing in this §5 obliges us to delete data that we are required to keep under binding legal authority (for example, a court order, tax authority request, or anti-money-laundering record-keeping rule). Where retention conflicts with a data-subject right, we will inform you of the conflict and the legal basis for retention.
6. Sub-Processors and Third Parties
To deliver the Service we rely on the sub-processors listed below. We enter into a written data-processing agreement with each sub-processor that processes personal data on our behalf. Each sub-processor is restricted to the processing instructions in the agreement and to the data categories required for its specific role. The Fortress posture means we prefer sub-processors established in Fortress-eligible jurisdictions; where a sub-processor is established elsewhere, we rely on standard contractual terms, the published DPA of the sub-processor, or an equivalent transfer mechanism.
- Hosting infrastructure (Postgres, Redis, application servers): DigitalOcean LLC. Data is stored in the region you select at signup (default: Frankfurt, Germany — DigitalOcean FRA1). DigitalOcean may incidentally receive personal data in operational metadata (for example, the source IP of a connection); that processing is governed by DigitalOcean's published DPA. Because Frankfurt is in the European Union, the German BDSG applies directly to the hosting layer; see §7 for the resulting transfer analysis.
- Transactional email (account verification, password reset, billing receipts, security alerts): Brevo (Sendinblue GmbH, Germany), reached over SMTP / STARTTLS on port 2525. Brevo receives the email address and the transactional content of each message. We have evaluated the Fortress posture for Brevo: their Germany establishment provides substantive data-protection guarantees under the German BDSG and the EU's GDPR — these are above and beyond what we require of sub-processors, and that is intentional.
- Payment processing, subscriptions, and tax: Polar.sh (Polar Software Sweden AB), reached over HTTPS. Polar is the merchant of record for paid plans. Polar returns to us only a customer-ID, subscription-ID, billing email, and tax identifiers where required by law. Polar is established in Sweden; we accept the residual cross-border transfer risk under Fortress's Sweden allowance.
- Slack integration: Slack Technologies, LLC (United States), only when you explicitly connect a workspace via OAuth. We receive only the scopes you authorise (channels:read, chat:write, and similar). Where we post an alert to a Slack channel on your behalf, the alert body may contain Customer Data (for example, a redacted decision summary); see §4.
- OAuth identity providers (sign-in with): GitHub Inc. and Google LLC (United States). On your first sign-in we receive your provider user ID, display name, and email address. You can revoke NullRun's access from your provider's authorised-apps page at any time.
- Database backups (Postgres
pg_dump+gzip, daily, with an automated restore test on every run): the snapshot is written to the production host, in the same region as the live database. An off-site copy to S3-compatible object storage (AWS S3 in prod) is supported by the backup tooling and can be enabled by ops; it is not part of the daily run today, so no data reaches a storage provider. See §5.
IP geolocation is not a sub-processor. The Fortress geo-block loads the MaxMind GeoLite2-Country database at the application edge and resolves the requesting IP to a country server-side, in order to decide whether the request comes from a sanctioned or otherwise high-risk jurisdiction. The lookup result is never transmitted to MaxMind, no Customer Data is sent to MaxMind, and MaxMind therefore does not process Customer Data and does not appear in the sub-processor list above.
Changes to sub-processors. We will notify you of any material change to this list (an addition or replacement of a sub-processor that processes Customer Data) by email and (where appropriate) by an in-product banner at least 30 days before the change takes effect, so that you have a reasonable opportunity to object. A non-material change (for example, a sub-region move within the same sub-processor, or a change that does not affect data categories or transfer locations) does not trigger the 30-day window.
We do not sell personal data, and we do not share it with third parties for their own purposes.
7. International Data Transfers
NullRun stores data in the region you select at signup (see §8 below). Where personal data is transferred across borders, we rely on the published DPA of the recipient, the recipient's standard contractual terms, or another lawful transfer mechanism accepted under the law of the customer's jurisdiction.
Frankfurt (Germany) and the European Union. The default data residency is Frankfurt (Germany). Because Frankfurt sits in the European Union, German BDSG applies directly to the hosting layer for any EU-located former customer or residual traffic (probes, partner-managed endpoints, or retrospective deletion requests). The Fortress geo-block at the edge prevents EU/EEA/UK residents from creating accounts, so the GDPR is not engaged as a controller regime under Art. 3(1) (no establishment of the controller in the Union) or Art. 3(2) (no targeting of EU data subjects) for the bulk of customer traffic. We nonetheless publish this Policy, our DPA, and our security posture so the access path is unambiguous if an EU authority ever inquires about residual traffic. Where a transfer occurs from a customer's home jurisdiction to Frankfurt, we rely on the recipient's published DPA or standard contractual terms.
United States sub-processors. Some of our sub-processors (Slack, GitHub, Google) are headquartered in the United States. We acknowledge that US national-security laws may permit US authorities to compel disclosure of personal data held by those sub-processors. We address this risk in the following layered way: (a) we rely on the published DPA of each US sub-processor; (b) we apply the security measures in §9 (encryption in transit and at rest, segregated keys, application-layer per-tenant scoping on every tenant-scoped query) to reduce the exposure of plaintext personal data to compelled disclosure; and (c) where a US sub-processor offers a regional data-residency option that lets us keep customer data in a non-US region, we use that option (see §8).
No transfers to comprehensively-sanctioned jurisdictions. We do not transfer personal data to, or store it in, any comprehensively-sanctioned jurisdiction. The canonical list of such jurisdictions is published in §16 below; we update that list (and this §7 by reference) when the underlying sanctions posture changes.
8. Data Residency
NullRun stores production data in the region you select at signup. The default is Frankfurt, Germany (DigitalOcean FRA1); customers may select another Fortress-eligible region at signup, and the choice is recorded against the organisation.
Best-effort commitment. We use commercially reasonable efforts to keep active production data within the region you selected. This commitment covers the primary database, primary Redis cache, and the application servers that serve your traffic.
Limited exceptions. The following are documented exceptions to the region guarantee and do not, individually or in aggregate, change the best-effort characterisation of this §8:
- Disaster-recovery failover. Where a primary region experiences a sustained outage, we may fail over to a backup region. We will restore primary-region operation as quickly as reasonably practical and notify you where the failover is sustained (more than 24 hours) or where it materially changes the data-residency profile of the Service.
- Backup snapshots. The daily snapshot is written to the production host, in the same region as the primary database, and is overwritten on a 7-day cycle. An off-site copy to object storage can be enabled by ops; it is not part of the daily run today.
- Sub-processor processing. Some sub-processors listed in §6 (for example, email, payment, sign-in, alert delivery) process data outside the region you selected. Those transfers are governed by §7 above and by the relevant sub-processor DPA.
- Security and abuse investigation. Where a security or abuse-investigation need requires it, NullRun personnel or a designated security partner may access operational data from a region other than the one you selected. Access is logged, restricted to a need-to-know list, and governed by the security measures in §9.
- Legal process. Where a binding legal process requires data to be produced in a specific jurisdiction, we will comply with the order (after, where permissible, notifying you and giving you an opportunity to challenge it).
If you require a contractual data-residency guarantee stronger than best effort, please contact us to discuss an enterprise addendum with stronger data-residency terms.
9. Security Measures
We apply technical and organisational measures designed to protect personal data against unauthorised or unlawful processing, accidental loss, destruction, or damage. These include:
- TLS 1.3 for all client connections and internal service-to-service calls.
- AES-256 encryption of sensitive columns at rest (Slack OAuth tokens, two-factor secrets, billing metadata where required), using PostgreSQL's pgcrypto with keys stored outside the database tier.
- Argon2id for password hashing with per-password salts.
- Application-layer per-tenant scoping on every tenant-scoped query (formal Postgres Row-Level Security policies are an active refactor — see
backend/src/db/REFACTOR_BACKLOG.mdstep 6; the outbox processor currently runs as a BYPASSRLS role, audited separately). - Hash-chained, append-only audit log so any retroactive tampering is detectable.
- PII redaction at the logging boundary: email addresses in log lines are replaced with a SHA-256 prefix (12 hex chars) so operators can correlate events for one user without the plaintext address landing in log storage.
- Cookie hardening: production cookies use the
__Host-prefix, theSecure,HttpOnly, andSameSite=Laxattributes as applicable, and a 7-day session lifetime aligned with our Redis session store. - CSRF double-submit protection on all state-changing browser requests; Bearer and SDK API-key requests are exempt because browsers do not auto-attach those headers to cross-site requests (CORS preflight blocks them).
- Rate limits, anomaly detection, and fail-closed defaults on enforcement paths so that a transport error does not silently widen budget or quota gates.
- Restricted employee access to production data on a need-to-know basis, with access logged.
- Fortress geo-block at the application edge: requests from ineligible jurisdictions are rejected before they touch the application server.
- Sanctions screening on signup: where the name/email screening layer has been enabled for the deployment, account-name and account-email screening against the OFAC, EU, UK, and UN consolidated lists is applied on top of the always-on IP-level Fortress geo-block.
No system is perfectly secure. Where a security incident affects your personal data, we will notify you and any competent authority in accordance with §10 and the applicable law in your jurisdiction.
10. Personal Data Breach Notification
We treat a personal-data breach as a confirmed event where there is a reasonable basis to believe that unauthorised access to, or unauthorised disclosure of, personal data has occurred. Suspected events are escalated to the security team and triaged before we make a notification decision.
Timeframes. Where a breach affects personal data of customers in a jurisdiction with a statutory notification window, we will notify the relevant authority without undue delay and, where feasible, no later than the window required by that jurisdiction's law. The default target is 72 hours from the point we become aware of the breach — the clock does not wait for severity classification, because the statutory windows do not — and it is consistent with the most common statutory window (Singapore PDPA, Turkey KVKK, Brazil LGPD, etc.) and the over-and-above EU GDPR / UK GDPR standard that we follow voluntarily where the breach also affects EU / UK residents in our incidental flow.
Severity classification. We classify breaches into three levels based on the categories of data affected, the volume of data subjects, the reversibility of the harm, and the mitigations available:
- High. Sensitive personal data (under the laws referenced in §16), authentication credentials, payment data, or a large-volume breach (thousands of records). Notification to data subjects is presumed; the competent authority is notified within 24 hours where feasible.
- Medium. A limited set of identifiers or operational metadata (for example, source IP and user agent without content), or a low-volume breach affecting a small set of users. The competent authority is notified within 72 hours; data-subject notification is considered on a case-by-case basis.
- Low. A reversible event where no personal data is exposed beyond what is already public, or where redaction and re-issuance make harm unlikely. We log the event internally; we do not notify the competent authority unless required by law.
Customer notification. Where you are a NullRun customer and the breach affects your Account Data, Billing Data, Customer Data, or Operational Data, we will notify you by email to the address associated with your organisation's primary contact (and, where appropriate, by an in-product banner) without undue delay and, where feasible, no later than 24 hours after we have classified the breach as High or Medium. The notification will include the categories of data affected, the likely consequences, and the measures we have taken or propose to take.
Documentation. We document every confirmed breach in our internal breach register, including the facts, effects, and remedial actions. We make that documentation available to the competent authority on request.
11. Your Rights
Subject to applicable law, you have the rights described below. We will respond to a verified request within the time period required by law (typically 30 days in most Fortress-eligible jurisdictions, extendable once where the law permits and the request requires additional time).
- Access: confirm whether we process your personal data and obtain a copy.
- Rectification: correct inaccurate or incomplete personal data.
- Erasure: delete your personal data, subject to the retention exceptions in §5.
- Restriction: limit how we process your personal data while a complaint is investigated.
- Portability: receive personal data you provided to us in a structured, commonly used, machine-readable format.
- Objection: object to processing based on our legitimate interests.
- Withdraw consent: where processing is based on consent, withdraw it at any time without affecting prior lawful processing.
- Non-discrimination: we will not deny service, charge different prices, or provide a different level of quality because you exercised any of these rights.
- Lodge a complaint with the competent data-protection authority in your country of habitual residence or place of business. The list of authorities we recognise is published in §16. We would, however, appreciate the chance to address your concerns directly first.
You can exercise most of these rights directly from the control center (profile, security, organisation, billing tabs) or by emailing support@nullrun.io with "Privacy request" in the subject line. We may need to verify your identity before acting on a request to prevent unauthorised disclosure.
13. Children
The Service is not directed to children under the age of 18 (or such higher age as applies in your jurisdiction). We do not knowingly collect personal data from children. If you believe a child has provided us with personal data, please contact us at support@nullrun.io and we will delete it promptly.
14. Cross-Border Data Flows and APEC CBPR
Where NullRun transfers personal data across borders in the Asia-Pacific region, we work towards the Asia-Pacific Economic Cooperation ("APEC") Cross-Border Privacy Rules ("CBPR") system. The APEC CBPR system provides a framework for organisations to ensure the protection of personal data transferred among participating APEC economies. Until full certification, we honour the substantive APEC CBPR principles (preventing harm, notice, collection limitation, use of personal information, integrity, security safeguards, access and correction, accountability) in the cross-border transfer context.
We do not currently hold APEC CBPR certification. Customers who require APEC CBPR certification as a procurement condition should request an enterprise addendum.
15. Data Processing Addendum (DPA)
Where NullRun processes Customer Data on your behalf as a processor, the parties' data-protection obligations are set out in our Data Processing Addendum (the "DPA"). The DPA is incorporated into our Terms of Service by reference once accepted (for online acceptance, the "Accept DPA" action in the control center or at checkout; for enterprise customers, the signed order form).
The full DPA is published at /dpa. Customers can record acceptance of the current DPA from the control center (organization page, Compliance section) or via the checkout flow; each acceptance is recorded against the org and emitted to the hash-chained audit log.
16. Applicable Law and Eligible Jurisdictions
The Service is offered to customers in the following jurisdictions. The list mirrors the canonical list in the Terms of Service §16; the Terms are authoritative where the two diverge, and we update both on a material change.
CIS (8): Armenia, Georgia, Kazakhstan, Kyrgyzstan, Uzbekistan, Tajikistan, Azerbaijan, Moldova.
GCC and Middle East (12): Turkey, United Arab Emirates, Saudi Arabia, Qatar, Bahrain, Oman, Kuwait, Egypt, Morocco, Tunisia, Jordan, Lebanon.
Africa (15): Nigeria, Kenya, South Africa, Ghana, Senegal, Ethiopia, Tanzania, Uganda, Rwanda, Côte d'Ivoire, Zambia, Zimbabwe, Botswana, Mauritius, Mozambique.
Latin America (15): Mexico, Brazil, Argentina, Chile, Colombia, Peru, Uruguay, Paraguay, Costa Rica, Panama, Ecuador, Bolivia, Dominican Republic, Guatemala, Honduras.
North America (2): United States, Canada(added 2026-09-16 — CCPA / CPRA thresholds not met at the pre-revenue stage; see the Fortress change note on geo_block.rs).
Asia-Pacific (15): Japan, South Korea, Singapore, Hong Kong SAR, Taiwan, Vietnam, Thailand, Malaysia, Indonesia, Philippines, Australia, New Zealand, Israel, Sri Lanka, Bangladesh.
Other (4): Serbia, Bosnia and Herzegovina, Montenegro, Albania.
Ineligible jurisdictions. The European Union, the United Kingdom, the European Economic Area, Switzerland, the People's Republic of China, the Republic of India, the Russian Federation, the Islamic Republic of Iran, the Democratic People's Republic of Korea, the Syrian Arab Republic, the Republic of Cuba, the Republic of Belarus, the Crimean / Donetsk / Luhansk regions, the Republic of Venezuela, the Republic of the Union of Myanmar, and the Islamic Republic of Afghanistan. The Fortress geo-block at the application edge rejects requests from these jurisdictions. The same ineligible list governs the cross-border-transfer prohibition in §7 above; that section defers to this list for the canonical enumeration.
Personal-data territorial carve-out. The contract governing law is the Republic of Armenia (Terms of Service §20). Personal-data processing, however, is governed by the law of the data subject’s habitual residence or the customer’s place of business — including, where applicable, the EU General Data Protection Regulation, the UK GDPR, the Swiss FADP, the California CCPA / CPRA, and comparable instruments — by virtue of their territorial scope, and cannot be displaced by a contractual choice-of-law clause. Where the law of the customer’s jurisdiction grants rights or imposes obligations beyond what is set out in this Policy, those rights and obligations prevail over the conflicting term.
Recognised competent authorities. Customers may lodge a complaint with the data-protection authority in their country of habitual residence or place of business. The following authorities are recognised by NullRun: Armenia — Personal Data Protection Agency under the Ministry of Justice; Turkey — Kişisel Verileri Koruma Kurumu (KVKK); Brazil — Autoridade Nacional de Proteção de Dados (ANPD); Japan — Personal Information Protection Commission; South Korea — Personal Information Protection Commission; Singapore — Personal Data Protection Commission; UAE — Data Office under the Telecommunications and Digital Government Regulatory Authority; Saudi Arabia — Saudi Data and AI Authority (SDAIA); Israel — Privacy Protection Authority; Australia — Office of the Australian Information Commissioner (OAIC); New Zealand — Office of the Privacy Commissioner; United States — Federal Trade Commission (FTC), with concurrent jurisdiction in the relevant State Attorney General and the offices of State-level privacy regulators (California Privacy Protection Agency under the CCPA / CPRA, Virginia VCDPA, Connecticut CTDPA, Colorado CPA, etc.); Canada — Office of the Privacy Commissioner of Canada (OPC) at the federal level, with concurrent jurisdiction of the Commission d’accès à l’information (CAI) for residents of Québec; Hong Kong SAR — Office of the Privacy Commissioner for Personal Data; Mexico — Instituto Nacional de Transparencia, Acceso a la Información y Protección de Datos Personales (INAI); Argentina — Agencia de Acceso a la Información Pública; Kazakhstan — Committee for Information Security of the Ministry of Digital Development, Innovations and Aerospace Industry. The list is non-exhaustive and is updated when the relevant law changes.
17. Changes to This Policy
We may update this Policy from time to time. When we do, we will revise the "Last updated" date at the top of this page. If a change is material, we will notify you by email and (where appropriate) by an in-product banner in advance of the change taking effect, where reasonably practical and where required by applicable law, so that you have time to review the change and, if you do not agree with it, close your account before it applies. Continued use of the Service after the effective date constitutes acceptance of the revised Policy.
18. Contact
For any question, complaint, or rights request relating to this Policy, please contact us. The data controller for the processing described in this Policy is:
Individual Entrepreneur Anatolii Maltsev
Country of establishment: Republic of Armenia
Tax identification: 20275845 (TIN / public service number)
Email (data-protection matters): support@nullrun.io
Email (general support): support@nullrun.io