AIRI loci Privacy Policy

United States version | Counsel review draft October 22, 2026

Operator: HongKong AlRl Technology Limited

Registered address: Room 803B, 8/F., West Coast International Building, 290-296 Un Chau Street, Cheung Sha Wan, Kowloon, HONG KONG

Registration number: 77292810

Last updated: October 22, 2026

Effective date: October 23, 2026

HongKong AlRl Technology Limited ("AIRI loci," "we," "us," or "our") operates AIRI loci. Our registered address is Room 803B, 8/F., West Coast International Building, 290-296 Un Chau Street, Cheung Sha Wan, Kowloon, HONG KONG. This Policy explains how we collect, use, disclose, retain, and protect personal information in connection with the US service environment of AIRI loci, including the website, desktop modeling application, workspaces, projects, pages, Agent, and related Services, and how you can exercise your rights. We use your IP address to identify an initial service environment among CN, US, and EU, and you may select the US or EU environment. An account and a workspace are assigned a home region when created. Under the current service design, that assignment is fixed, and subsequent requests for the account or workspace are routed to its assigned region. Changing the region selector or accessing the Services from another location does not automatically migrate an existing account, workspace, or its content, or create a copy of the account in another region. Region selection does not change your place of residence or waive rights under mandatory applicable law.

1 Scope and our role

This Policy covers personal information processed through the Services and related support and billing interactions. Additional feature-specific or regional notices apply where provided; they control for the matters they expressly address. It does not govern an independent third-party service merely because a link to it appears in AIRI loci.

Our Third Party Information Sharing List accompanies this Policy and describes relevant categories, purposes, sources, and recipients. It should be read together with this Policy; it does not expand our permitted uses of information.

We generally determine the purposes and means of processing account, billing, security, and service-administration information. Where an organization determines the purposes and means for personal information in its content and engages us to process it under a data processing agreement or other written instructions, our role and obligations for that processing are governed by that arrangement and applicable law. Contact the relevant organization for requests it controls; we will assist as required and will not refer away requests concerning processing for which we are responsible.

A workspace manages projects, members, permissions, usage, and storage. A project organizes pages, uploaded materials, and project memory. A page is an expandable creative canvas. Assets include supported text, files, images, audio, video, HTML, and other materials. User Content includes your inputs, instructions, uploads, generated results, and related saved content.

2 Information we collect and why

Accounts and authentication

You may sign in with Google, Microsoft, or an email one-time passcode (OTP). We process your email address, account identifiers, and the verification or authentication information needed to create, authenticate, and secure your account. For email OTP, we process the destination address and delivery and verification information needed to send and validate the code. You may provide a nickname and avatar. Basic registration does not require a company, job title, industry, or name unless a specific feature clearly explains otherwise.

When you choose Google or Microsoft sign-in, we receive the account identifiers and profile information supplied under the permissions you authorize in the sign-in flow. We use this information to authenticate you and maintain your account. This sign-in authorization does not give AIRI loci general access to your email messages or cloud files.

Workspace and content information

We process workspace, project, and page names, creators, member relationships, permissions, timestamps, and settings to organize your work and support collaboration. We collect and process content you submit or upload, including prompts, text, images, audio, video, documents, design materials, and other supported files, together with results produced from your instructions. We use this information to perform tasks, call needed models and tools, return and save results, support editing and reuse, and carry out your requested operations.

Other users may include information about you in their content or invitations. We may also receive information from authorized sign-in providers, collaborators, payment providers, and other service providers as necessary to provide the relevant feature. We handle this information consistently with the disclosed purposes and applicable law.

Personal memory and project memory

Personal memory is associated with your account. Among users of the product, only you may view your personal memory; other users and workspace administrators may not. You may disable, edit, or delete it. Disabling it stops new personal-memory additions and its use in subsequent tasks. Manually deleted personal memory cannot be recovered through the product. Authorized operational access and lawful retention exceptions are described in Sections 3 and 7.

Project memory supports continuity in a project. Only a user who is a workspace super administrator, administrator, or member and also has editing or management permission for that project may view, upload, edit, disable, or delete project memory. Disabling it stops new additions and its use in subsequent tasks. Directly deleting it is not recoverable through the product. If a project is deleted instead, associated project memory follows the project's 30-day recovery period and may be restored with it.

Collaboration and sharing

We process account details, memberships, permissions, and relevant activity records to display content to authorized collaborators. Invitation links can be successfully accepted once, expire as shown, and require the recipient to register or sign in. Read-only recipients may view only the content their permissions allow. AIRI loci does not currently offer an internet-wide public community or publicly indexed projects; a "public project" is open only to appropriately authorized workspace members.

Workspace super administrators can view and manage all projects in their workspace. The super administrator and project creator are fixed project managers, subject to supported succession processes. When a member leaves or is removed, their nickname and avatar cease to be displayed in that workspace and are replaced by "Former member." Their team content is managed by the super administrator or another supported successor. These management arrangements do not automatically transfer intellectual property rights or remove statutory privacy rights.

Device network usage and task records

We process necessary information about IP addresses, browser and operating system, sessions, request identifiers and paths, processing status, duration, errors, model or workflow calls, and credit consumption to operate the Services, protect security, troubleshoot failures, and verify billing. Request and response records may contain prompts, attachment references, model parameters, Output, and provider information. We apply masking, truncation, or other safeguards to passwords, access tokens, private keys, and credential-bearing links.

Payments subscriptions and billing

We process workspace and plan identifiers, credit and storage usage, order and transaction identifiers, amounts, currency, payment time and status, refunds, recurring-payment authorization and cancellation, renewal attempts, and subscription status. This supports payment, entitlement delivery, reconciliation, fraud prevention, customer support, and required financial records.

Payments for the US service are processed through Stripe. You provide the payment information requested in the Stripe payment flow, and we exchange the order, payment, subscription, refund, and reconciliation information needed to administer your purchase. Stripe also processes information for its own payment, fraud-prevention, and legal-compliance purposes as explained in the privacy information provided in its payment flow. We provide additional recipient information where required by applicable law.

We process renewal records to avoid duplicate charges and stop subsequent attempts after payment succeeds. Deleting a workspace stops future automatic renewal; recovery does not automatically restore authorization. For receipt or invoice requests, we process necessary order, billing, tax, and delivery details. The current product has no self-service invoice feature.

Support and communications

When you contact us by email or through our Discord community, report an issue, give feedback, submit a complaint, or exercise a privacy right, we process the contact details, messages, attachments, and information reasonably necessary to verify and resolve the request. Your use of Discord is also subject to Discord's own terms and privacy policy. We currently do not conduct email marketing and do not use customer content or projects for public case studies or promotion. Authentication, transaction, security, and requested support messages are service communications. If we introduce email marketing, we will first provide appropriate disclosures and any legally required choice or consent and an effective unsubscribe method.

3 Agent processing and model training

To perform a task, an Agent may access relevant conversations, the current project and page, selected assets, uploaded materials, personal memory or project memory where enabled, and other context available within your permissions. The desktop modeling application processes the files, modeling instructions, application context, and results needed for the operations you authorize on your device. Local execution does not mean the whole feature is offline: necessary task content may be sent to AIRI loci and model or tool providers, and deliverable results may be uploaded for saving and use in AIRI loci. Access is limited by the task authorization and permissions actually granted. The Agent does not currently act for you to send messages, publish, purchase, or otherwise operate external online accounts.

We use model and tool service providers to carry out your generation and other requested tasks. For this purpose, we process and disclose the necessary instructions, relevant conversation context, reference materials, and files or accessible file links, and receive and store the resulting content for your use. A provider given an accessible file link may retrieve the file itself. The providers involved depend on the model or tool used for the task. Recipient categories and purposes are described in Section 5, US data residency and remote operational access in Section 6, and retention and deletion in Section 7.

AIRI loci does not use User Content, including inputs, uploaded files, workspaces, projects, pages, assets, Agent conversations, memories, or generated results, to train AI models. We may use necessary technical and usage information to maintain performance, troubleshoot, and improve the product without using User Content for model training.

We do not permit model or tool providers processing User Content on our behalf to use that content to train AI models. They may process the content necessary to perform your tasks and for the limited operational, safety, or legal purposes applicable to their services, subject to our commitments and applicable law. The no-training commitment does not mean zero retention: task execution, permitted safety review, temporary processing, and legally required records may involve retaining information. We limit such processing through the applicable service arrangements and provide additional notices or obtain consent where required.

We may automatically inspect necessary instructions, uploads, results, and records for unlawful content, fraud, abuse, security risks, or failures. A limited number of authorized personnel subject to confidentiality obligations may access original content only for support requests you initiate, a specific fault, security or abuse investigations, or legal obligations. Access is controlled and logged. Such access does not authorize model training by AIRI loci.

Automated processes may manage concurrency, usage limits, credit reservation and settlement, duplicate-payment prevention, and security detection. Payment or refund exceptions may undergo human review. Personal memory may personalize Agent responses, but AIRI loci does not currently use it for personalized marketing or profile-based differential pricing, and does not permanently ban accounts solely through these automated administrative processes. Where applicable law grants rights concerning significant automated decisions or profiling, Section 9 explains how to contact us.

4 Cookies usage analytics and browser privacy signals

We use necessary cookies, local storage, and session storage for authentication, language and feature preferences, restoring an Agent conversation, request continuation, and security. A language cookie may last up to one year. Login cookies persist until the relevant session expires, you sign out, or browser data is cleared. Other preferences generally remain until cleared, reset, or no longer needed; short-lived session information expires with its session or configured lifetime.

We analyze existing account, billing, workspace, and recorded Canvas activity to understand product use, activity and retention, service performance, credit consumption, and revenue. The records include user and workspace identifiers, timestamps, plan information, member counts, sign-in source, metered usage, and recorded Canvas operations. We also process model-call and Agent-operation metadata, including user, workspace, Canvas and session identifiers, model, start and end times, status, and token usage, for metering, billing, reliability, and troubleshooting. We handle linkable usage records as personal information. Content processing for conversations, saved materials, and model calls is described in Sections 2 and 3. Before introducing materially different collection or use, we provide the notices and choices required by applicable law.

You may use browser settings to clear or restrict cookies and local storage, although this may affect authentication, saved preferences, and other necessary functions. AIRI loci does not currently automatically change its processing in response to browser Do Not Track (DNT) or Global Privacy Control (GPC) signals. You may submit privacy requests to contact@airilab.com as described in Section 9. This description of the current technical functionality does not limit rights under applicable law. Clearing browser storage does not itself delete information retained in your account or our service systems.

5 How we disclose personal information

We disclose personal information only for the described purposes and as permitted by applicable law. Recipients may include hosting, storage, content-delivery, model, search, file-processing, authentication, payment, communication, security, and support providers. Providers acting on our behalf are subject to appropriate contractual restrictions and safeguards. Where another party independently determines its purposes and means, we provide the disclosures and obtain consent required for that relationship; an independent provider's policy does not replace our own obligations.

We also disclose information to collaborators and recipients you or authorized administrators select, within the relevant permissions; to professional advisers as needed under appropriate obligations; in a lawful merger, acquisition, financing, restructuring, or asset transfer subject to applicable protections; and to authorities or other parties where legally required or reasonably necessary and lawful to address fraud, security, legal claims, or threats to rights and safety. We do not make project content public to the internet through a public-community feature.

The recipient categories in this section are used for the functions described in Sections 2 and 3: infrastructure providers support hosting and storage; model, search, and file-processing providers perform requested tasks; authentication and communication providers support sign-in and service messages; payment providers process purchases; and security and support providers help protect and operate the Services. The information disclosed is limited to what is needed for the relevant function. Where applicable law requires us to identify a particular recipient or provide a list of specific third parties, we provide that information through the relevant notice or in response to a valid request, as required. Where we process information for an organization, any additional subprocessor disclosure and notification obligations are governed by the applicable data processing agreement. We provide additional notice or obtain consent for changes where required.

6 US data residency and remote operational access

Our primary servers for the US service environment are hosted on Amazon Web Services (AWS) in Oregon, United States. We host and store service data in the United States, including account and workspace records, User Content, service logs, and backups. Processing performed by AIRI loci and providers acting on our behalf to execute tasks takes place in the United States. We do not move these data outside the United States for hosting, storage, backup, or task execution. Authorized remote operational access from outside the United States is described below.

Authorized employees located outside the United States, including in China, may remotely access the US service environment only to perform necessary maintenance, troubleshoot specific faults, or investigate security incidents. Access is limited to the information necessary for the assigned work, subject to access controls, confidentiality obligations, and access logging. These employees may not download or retain service data outside the United States for operational work. Remote access may involve viewing or otherwise processing information from abroad; the US hosting and storage commitment does not mean that all personnel accessing the data are located in the United States.

This commitment concerns the service environment operated by AIRI loci and task processing carried out on our behalf. It does not describe information you provide directly to independent sign-in, payment, or community services for their own purposes, or copies you or authorized collaborators choose to download or share. Relevant independent services and data flows are described in our Third Party Information Sharing List and their own privacy notices; their involvement does not create an exception for our own US hosting, storage, backup, or task-execution activities.

7 Retention deletion and recovery

We retain personal information only as long as reasonably necessary for the purposes described, taking account of active use, task completion, your choices, security, billing, tax, audit, dispute resolution, and applicable legal requirements. We do not treat every category as having the same retention period.

Deleted Canvas content and elements covered by the recycle bin have a 30-day recovery period. After that period, scheduled background processing identifies expired records and deletes the Canvas and its associated operation records, Canvas snapshots, and recycle-bin records, and releases the associated asset references. An asset that remains referenced by another Canvas or conversation is retained for that use; assets with no remaining references enter physical cleanup. Physical cleanup follows the scheduled process and is not necessarily completed at the instant the recovery period expires. Directly deleted personal or project memory cannot be recovered through the product. These storage rules do not override applicable statutory deletion rights; backups and lawful retention are addressed below.

Closing an account immediately disables it and starts a 30-day period during which support may restore it. During recovery, information is not used as if the account were normally active; use is limited to recovery, necessary security, and legal obligations. After recovery expires, we delete or de-identify the relevant personal information subject to lawful exceptions. A closed account or departed member appears as "Former member" in the relevant workspace; team content may remain with the organization according to permissions, applicable law, and the nature of the request.

A deleted workspace has no self-service recycle-bin recovery; contact support within 30 days to request recovery before business-content deletion or de-identification begins. These product recovery periods do not override applicable statutory deletion rights. Privacy requests are assessed separately, including information in backups and information held by relevant processors where required.

We may retain limited information where necessary and lawful for transactions, taxes, security, legal claims, or other legal obligations. Backup copies are removed through the applicable rotation and deletion process. Pending removal, retained backup information is protected and restricted from ordinary service use, except as necessary for disaster recovery, security, or legal obligations. When a backup is restored, applicable deletion requests must continue to be respected. Retention is determined by the purpose of the record, the recovery and backup process, and applicable legal requirements; it does not extend automatically to all information associated with an account.

Operational and security logs, payment and tax records, support requests, and model-provider records are retained for the periods necessary to meet their service, security, accounting, dispute-resolution, or legal purposes. When AIRI loci provides an export without a visible AI label, our product policy is to retain the account, content identifier, request time, agreement version, result, and necessary traceability records for six months, or longer only where legally required or otherwise justified by a disclosed lawful purpose. This six-month product retention rule is not presented as a uniform requirement of United States law and remains subject to applicable deletion rights and lawful exceptions.

If the Services cease, we stop new collection for those Services, provide required notices, and delete or de-identify information except for lawful retention. Where immediate deletion is not technically possible, retained data remains restricted and protected until deletion can be completed in accordance with applicable law.

8 United States state disclosures

The additional rights and disclosures in this section apply where a relevant United States state privacy law covers you, our activities, and the information concerned. The scope of rights and applicable exceptions varies by state.

We collect identifiers and account details; commercial and billing information; internet, device, and network activity; user-submitted text and audiovisual content; communications with support; and preferences or contextual information you provide or enable in memory. Sources include you, your device, authorized collaborators, sign-in providers, and transaction and technical providers. Purposes and recipient categories are described in Sections 2 through 5, and retention in Section 7. We describe collection before or at the point of collection where required, including any new sensitive-data processing.

The product currently contains no advertisements, and we do not currently conduct email marketing or use customer content or projects for public case studies or promotion. We do not use personal memory for personalized marketing or profile-based differential pricing. Our current usage analysis serves the service-administration and product-improvement purposes described in Section 4. Your applicable rights concerning disclosures and other processing are described in Section 9.

Uploaded content may contain sensitive information about you or others, even though basic use does not normally require it. We do not currently offer dedicated facial recognition, voiceprint identification, identity-document recognition, precise-location profiling, or medical-diagnosis features. Submit sensitive information only when necessary for a supported task and you have the required authority. Where law requires prior consent or a limitation mechanism for sensitive-information processing, we provide it; an upload is not treated as satisfying every legally required consent.

We do not currently use personal memory for personalized marketing or profile-based differential pricing. Any legal right to limit sensitive-information use or opt out of profiling applies according to the statutory definition and our actual processing, not simply because the product uses AI. Significant new uses will be disclosed and any required consent obtained before they begin.

9 Your choices and privacy rights

Depending on applicable law, you may request confirmation of processing; access to and a copy of personal information; correction; deletion; a portable copy; information about relevant disclosures or recipients; and an explanation of processing. You may also have rights to opt out of sale, sharing, targeted advertising, or profiling used for decisions producing legal or similarly significant effects, and to limit certain uses or disclosures of sensitive information. Some rights are subject to exceptions.

Email contact@airilab.com with the subject "Privacy request" to submit an access, correction, deletion, or other applicable privacy request. Existing content, memory, or permission controls may also be used for their supported purposes. We request only information reasonably necessary to verify identity or authority for the particular request. We do not impose verification on an opt-out request where applicable law prohibits doing so. You may use an authorized agent where permitted; we may request evidence of authority and identity verification as allowed by law.

Where the CCPA applies, we confirm receipt of requests to know, correct, or delete within 10 business days and ordinarily provide a substantive response within 45 calendar days. If a permitted extension is necessary, we notify you within the initial period and explain the reason. Opt-out, limitation, and other requests follow their own applicable deadlines. Elsewhere, we respond within the period required by the applicable state law, with any lawful extension and notice. We explain a refusal or partial refusal unless law prohibits disclosure.

If applicable state law gives you an appeal right, reply to our decision or email contact@airilab.com with the subject "Privacy appeal," explaining the decision challenged. We review the appeal and respond with reasons within the applicable statutory period. If the appeal is denied, we provide information on contacting the relevant state attorney general or regulator where required. We do not unlawfully discriminate or retaliate because you exercise a privacy right.

You may withdraw consent for processing based on consent; this does not undo lawful processing already undertaken. Some information remains necessary to deliver requested services, fulfill a contract, or meet legal obligations, and withdrawing or deleting it may affect the relevant feature. Workspace administrators cannot prevent you from exercising rights directly with AIRI loci for processing for which we are responsible.

10 Children and age eligibility

AIRI loci is primarily intended for adults, but individuals under 18 may use the Services with their parent's or legal guardian's agreement, guidance, and supervision as described in the Terms. Basic registration does not ask for age or date of birth; registration is not verification of age or guardian authorization. Where the Children's Online Privacy Protection Act (COPPA) applies, including when we have actual knowledge that we are collecting personal information online from a child under 13, we must provide the required notices and obtain verifiable parental consent before the covered collection, use, or disclosure, unless a specific lawful exception applies. General acceptance of these Terms or this Policy is not a substitute for that consent. A child may not use an affected feature until its legally required conditions are met. If we learn that required authorization is missing, we restrict the affected use and delete or otherwise handle the associated information as required by law. Parents or legal guardians may contact contact@airilab.com to ask about, review, or request deletion of their child's personal information, or to refuse further collection or use, subject to appropriate verification and applicable law. We also comply with other applicable protections for minors; parental permission does not waive those protections.

11 Security

We use technical and organizational safeguards appropriate to the information and risks, including applicable encryption, authentication, access controls, backups, monitoring, incident response, and personnel and provider management. No internet service can guarantee absolute security. Protect your credentials and configure workspace and sharing permissions carefully.

If a security incident triggers a notification duty, we investigate, take appropriate corrective action, and notify affected people and authorities in the manner and within the periods required by applicable law.

12 Changes to this Policy

We may update this Policy when laws, features, technology, or business practices change. We publish the updated version and date, and use product notices, email, or another conspicuous method to explain material changes before they take effect where required. If a new use requires consent, we obtain it separately. Publishing a new Policy does not itself authorize a materially different use of previously collected information contrary to prior commitments or applicable law.

13 Contact

Email contact@airilab.com for privacy questions, requests, complaints, or appeals. Our Discord community may provide general guidance, but this email is the formal privacy contact. We may ask for necessary information to identify the request and verify authority, and will explain any inability to respond where required.

Operator: HongKong AlRl Technology Limited

Registered address: Room 803B, 8/F., West Coast International Building, 290-296 Un Chau Street, Cheung Sha Wan, Kowloon, HONG KONG