Terms of Service

Originally published 26.07.21 · In case of any discrepancy, the Korean version prevails. View original

Last Updated: 2026-07-21

1. Purpose

The purpose of these Terms of Service ("Terms") is to set forth the rights, obligations, and responsibilities between Linkbricks Horizon-AI Inc. (the "Company") and Users in connection with the use of the Company's AI-based service "VORA" (the "Service") and its related features (personal AI creation, multi-agent conversation, automation tasks, file upload and RAG document analysis, AI tool subscriptions, friend invitations and AI sharing features, an admin console, and others).

Matters concerning the collection, use, storage, de-identification, and personalization of personal information that are not set forth in these Terms are governed by the Company's Privacy Policy.

"AI Tool" means an individual AI-based feature, provided by the Company or a third party, that a User may use through a subscription.

"Usage" means the number of calls, or an equivalent unit of use, deducted based on the successful generation of a result from use of an AI Tool.

An AI Tool's usage is managed under one of the following methods.
Call-Based usage (hereinafter "Call-Based")
Concurrent-task-capacity-based usage (hereinafter "Slot-Based")

"Friend AI" means the AI of another User to whom a User is connected through a friend relationship.

"My Storage Capacity" means the total storage limit for files that a User stores in the Service, such as for RAG document analysis, and consists of the capacity provided by default to all Users plus a paid expansion option that the Company may offer.

2. Effect and Amendment of the Terms

A User is deemed to have agreed to these Terms by registering as a member or by using the Service.
The Company may amend these Terms to comply with applicable law, improve the Service, or change its operating policy.
The Company will give notice of an amendment at least 7 days in advance; an amendment that materially affects a User's rights or obligations will be announced 30 days in advance.
A User who does not agree to amended Terms may discontinue use of the Service and delete their account; continued use of the Service is deemed acceptance of the amended Terms.

2-1. Language and Order of Precedence

These Terms may be provided in Korean, English, and Japanese.
In the event of any discrepancy in interpretation between the language versions, the Korean version shall prevail, provided that where mandatory law of the country in which a User resides provides otherwise, that law governs.

3. Account Creation and Authentication

An account may be created according to the procedure set by the Company (such as sign-up through an invitation link, or general sign-up separately permitted by the Company).
Sign-up is completed through a third-party OAuth authentication method such as Google, Kakao, or Line.
A User is responsible for maintaining the security of their own login information and account.
The Company is not liable for damages arising from theft or leakage of an account, except where the Company is at fault.
In signing up for and using the Service, a User must provide only their own information or information they are lawfully authorized to use, and all legal liability arising from the unauthorized use of another person's information is attributable to that User.
If a User disconnects VORA from a third-party OAuth provider (Google, Kakao, Line, etc.) used at sign-up, the Company will immediately expire that User's authentication token and proceed with the account withdrawal (account deletion) process. In this case, the 30-day grace period and data-handling procedure under Article 18 of these Terms apply in the same manner.

3-1. Minimum Age to Use the Service

Sign-up for and use of this Service is limited to persons who are at least 18 years old and who have reached the age of majority under the law of the country or region in which they reside (if the age of majority under local law exceeds 18, that age governs; for example, 19 in the Republic of Korea).

The Company may refuse a sign-up application from a User who does not meet this standard, or restrict or delete the account of a User already confirmed to have signed up in violation of it.

Where a User signs up using false information or another person's information, all liability arising therefrom is borne by that User.

The Company may request additional information where necessary to verify a User's age.

4. Processing of Personal Information

The Company processes a User's personal information in compliance with applicable law, and whether and to what extent personal information is collected, used, stored, destroyed, de-identified, used for service operation notices, statistically analyzed, or personalized is determined in accordance with the Privacy Policy.

4-1. Notice on Automated Processing and AI-Based Services

This Service may include automated processing for the purpose of providing AI-based features. Such automated processing is intended to perform Service functions and does not automatically make decisions that have a material legal or similarly significant effect on a User.

For chat data deleted by a User, or automatically deleted upon expiration of the retention period, the Company may retain such data for an additional period of up to 30 days from the time of deletion for the purpose of responding to requests to recover accidental deletion, responding to system failure or data loss, and preserving logs for legal proceedings, and will completely destroy it once that period elapses. Details are governed by the Privacy Policy.

5. Content of the Service Provided

The Service provided by the Company includes the following features.

Creating a personal AI agent and configuring its system prompt
An AI Tool-based subscription model (search, document generation, code generation, image generation, operations automation, etc.)
Inviting friends and sharing AI Tools (including configuring whether an invited friend may use one's AI)
Multi-agent chat and simultaneous tagging of up to 4 AIs
Power Research mode (a PDCA-based in-depth research feature)
Discussion mode (an argumentative debate feature among multiple AIs)
Creating and scheduling automation tasks (mailing, report generation, etc.)
File upload (RAG documents), image analysis, and storage features
Chat attachments: processed as temporary input for generating AI responses (not stored)
RAG document upload: storage of documents at the User's option and their use as reference material for AI responses
The storage limit for RAG documents may consist of a default provided capacity and a paid expansion option (My Storage Capacity Subscription); specific criteria are governed by in-service notices and the Pricing Policy.
Downloading generated AI output as a file (PDF, Word, PPT, etc.)
Exporting the entire content of a chat session (a feature to download the chat record in a format such as Word (DOCX) or Excel (XLSX))
An admin console (member management, revenue dashboard, notice management, etc.)
External sharing of conversation output

The Company may change, add, restrict, or discontinue all or part of the Service to improve quality, meet technical needs, or implement its operating policy.

6. User Content and Intellectual Property

6-1. Ownership of Input and Output

All data a User enters into the Service (documents, files, images, text, etc.) and the output generated by AI based on it belong to the User. The Company may use data and statistical metadata that have been de-identified or aggregated so that a User cannot be identified, solely for the purposes of improving Service quality, enhancing features, and ensuring stability; specific matters concerning the processing of personal information are governed by the Privacy Policy.

The Company does not use files, images, documents, text, chat content, or AI-generated output uploaded by a User to train the Company's own AI models, provided that the Company may carry out automated processing to the extent necessary to generate AI responses, perform RAG analysis, image analysis, and file processing, and perform Service functions requested by the User.

An image uploaded by a User may be used only for analysis in response to the User's request, for generating a response, or as a reference input for a supported AI feature, and is not used to train the Company's own AI models.

6-2. Responsibility for User Content

A User holds lawful rights to the content they enter into or upload to the Service, and is responsible for ensuring that such content does not contain high-risk sensitive information (resident registration numbers, passport numbers, account numbers, authentication credentials, medical information, etc.).

All legal liability and disputes arising from a violation of this are attributable to the User, and the Company bears no responsibility for such matters.

6-3. Nature of AI Output and Limits of Liability

AI output is generated by statistical and probabilistic methods and may contain inaccurate or incomplete information.

The Company does not guarantee the accuracy, completeness, or reliability of AI output. A User must use such output only as reference material and must not make decisions relying entirely on it.

A User may export (download) the entire content of a chat session within the Service in a document (DOCX) or spreadsheet (XLSX) format. Such an exported output is, for the User's convenience, a direct conversion of the content of the chat screen, and is not a document that the Company authenticates or officially certifies.

Because an exported output is provided in an editable format, a User may modify or process it after downloading, and the Company does not guarantee the authenticity or integrity of the output after download. All legal liability arising from a User forging or altering an exported output, or falsely presenting or distributing it as if it were the actual conversation content or an actual AI response, is attributable to that User.

6-4. Labeling of AI-Generated Content

In accordance with applicable law (including the Republic of Korea's Framework Act on the Development of Artificial Intelligence and Establishment of a Foundation for Trustworthiness), the Company may apply a label, watermark, or notice indicating that a result was generated by AI.
A User must not arbitrarily remove, alter, or conceal a label applied by the Company that indicates AI-generated content.
A User must not label or distribute AI-generated content in a manner that misleads others into believing it was created, filmed, or recorded by an actual person.

7. Prohibited Conduct

A User must not engage in any of the following acts.

Uploading, entering, requesting the processing of, or sharing another person's personal information or sensitive information on the Service without that person's express and lawful consent
Uploading, entering, requesting the processing of, or sharing a file, document, image, or text on the Service that contains high-risk sensitive information such as a resident registration number, passport number, account number, authentication credential (API key, password, etc.), or medical information
Generating or distributing content that is violent, promotes terrorism, hateful, discriminatory, self-harm-related, sexually exploitative of children, or obscene
Relying entirely on AI output for professional judgments in fields such as medicine, law, tax, or investment
Sending excessive requests to the system to circumvent rate limits or to threaten server stability
Unlawful acts such as distributing malicious code, phishing, spamming, fraud, or social-engineering attacks
Unauthorized access to or use of an AI model through reverse engineering, crawling, automated bots, or similar means
Uploading content that infringes a third party's copyright, trademark, trade secret, or other intellectual property right, or using such content to generate or use other content
Interfering with the normal operation of the Service or using the Service for an unlawful purpose
Uploading, entering, processing, or sharing an image, video, or audio recording containing the likeness, body, voice, or other personal identifying features of a living person other than oneself, without that person's express and lawful consent, or generating or synthesizing a new image, video, or audio recording using such material (including face swapping); a User bears the responsibility to directly obtain and be able to prove such consent
Impersonating another person, or creating an account or using the Service by using another person's name, email address, contact information, or account information without that person's consent
Entering another person's email address for authentication, receiving notifications, or an automation task (sending email, scheduled sending, sharing invitations, etc.) in a manner that may cause confusion, harm, or legal disputes for that third party
Using the Company's email-sending, automation-task, or scheduled-sending features to send bulk advertising, commercial, or repetitive messages without the recipient's prior consent, or to distribute spam, phishing, or fraudulent messages
Creating multiple accounts, generating abnormal call patterns, using external automation tools, or engaging in similar acts to circumvent or undermine the operating policy set by the Company (the "Service Operating Policy"), such as usage limits, storage limits, sending frequency, or automation-task execution criteria
Generating, for any purpose, synthetic images, video, or audio that is difficult to distinguish from an actual person (so-called deepfakes) without that person's express consent. The following acts in particular are strictly prohibited.

(a) Generating or distributing sexual or lewd synthetic content depicting an actual person
(b) Using the likeness, voice, or identity of a third party such as a celebrity, public figure, or entertainer to generate or distribute content that impersonates, misleads, spreads false information about, defames, sexually synthesizes, commercially misappropriates, or otherwise infringes the rights of that person, or that is likely to do so. Generating photorealistic synthetic content that could be mistaken for an actual person is not permitted, even for satire or parody purposes, without that person's express consent.
(c) Generating or distributing synthetic content for an unlawful or deceptive purpose such as defamation, fraud, identity impersonation, or the spread of fake news or political disinformation

However, where a User uses the likeness of themselves, or of a person over whom they hold lawful authority or express consent (such as a consenting friend) to generate an avatar or profile image, and the use does not fall under a prohibited purpose such as impersonation, deception, sexual synthesis, defamation, or the spread of false information, such use is permitted within the scope allowed by the Service.

Forging the origin or fact of generation of AI-generated content, or distributing it while concealing or removing the fact that it was AI-generated or a label applied by the Company indicating AI-generated content
Arbitrarily removing, altering, or concealing the unsubscribe link and related notice that the Company provides in automation-task (recurring or scheduled) emails, or circumventing that function

All legal liability and disputes arising from a violation of the preceding paragraph are attributable to the User, and the Company bears no responsibility for such matters, provided that the feature the Company provides for entering, forwarding, or sending to a third party's email address is available only where the data subject of that email address has given prior consent, and a User bears the responsibility to directly obtain and be able to prove such consent.

The Company provides an unsubscribe mechanism for emails sent through the automation-task feature. The Company merely provides the sending infrastructure for that feature and is not the sender; responsibility for the content of a sent email lies with the User who requested that it be sent.

7-1. Content Moderation, Reporting, and Action

To prevent the prohibited conduct under Article 7 and the generation or distribution of unlawful or harmful content, the Company may operate an automated content filtering and moderation system, together with the safety systems and content policies of the third-party AI model providers it uses.
A User who discovers unlawful, harmful, or rights-infringing content may report it to the Company through the in-service reporting feature or by email (contact@vora.im).
Where a report is received or a possible violation is detected by an automated system, the Company may review the relevant content within the minimum scope necessary for fact-finding, and, depending on the outcome, may take action under Article 18-1, such as deleting or blocking the content, restricting features, or sanctioning the account.
A holder of intellectual property rights, such as copyright, may explain the basis for an infringement claim and request that the Company delete or block access to the relevant content, and the Company will take necessary action in accordance with applicable law.
Where a possible violation is confirmed, the Company may retain the relevant content, account information, generation-request records, and logs within the necessary scope.
The Company will notify Users of ethical guidelines and legal compliance matters concerning the use of AI, through in-service notices, policy pages, and similar means.

8. Service Availability and Outages

The Company strives to provide a stable Service, but use of the Service may be temporarily suspended due to system maintenance, updates, security measures, or force majeure.
In the event of an outage, the Company will restore the Service as quickly as possible and will notify Users through a Service notice or another appropriate means.
The Company may add, change, or discontinue features according to the manner in which it provides the Service, its business policy, or technical changes.

9. Notice of the Limitations and Risks of AI

AI output may be inaccurate, distorted, biased, or contain hallucinations.
A User is responsible for verifying AI output, and the Company does not guarantee the accuracy, completeness, or suitability of such output.
A User must not make decisions based solely on AI output in a field that requires expert advice.
This Service uses third-party AI large language model (LLM) APIs, such as those of OpenAI and Anthropic, to generate responses. Responses to certain topics, expressions, or requests may be restricted or unavailable under the content policy or safety policy of the relevant model provider, which is a matter beyond the Company's control. The Company is not liable if a User does not receive a desired response as a result.
The Company does not arbitrarily disable, bypass, alter, weaken, or customize the content policy, safety policy, or moderation features provided by third-party AI model providers such as OpenAI and Anthropic. (The Company does not disable, bypass, weaken, modify, or customize the moderation or safety features provided by third-party AI model providers, including OpenAI and Anthropic.)

9-1. Restricted Administrator Access and No-Review-of-Conversations Principle

Except to the extent necessary for providing this Service, maintaining security, responding to incidents, and fulfilling obligations under applicable law, the Company does not view a User's AI conversation content or message records.

Conversation content is processed in an environment where administrator access is technically restricted, and is processed only in an automated manner to provide the Service and to perform functions requested by the User.

However, exceptional access or processing within the minimum necessary scope may occur in the following cases.

Automated content filtering, moderation, and safety review (Article 7-1)
Fact-finding in response to a User report or an automatically detected possible violation
A response required to fulfill an obligation under applicable law or to comply with a lawful request from an investigative authority

10. Limitation of Liability

The Company is not liable for damages arising from any of the following.

The accuracy, completeness, or suitability of AI output
Damages arising from the use of AI output or an automation result
A User's input data or the result of its use
Damages arising from a User's failure to manage or configure their account properly
A User's decision based on the result of automated processing
Service interruption, or data loss or damage
Automatic deletion of data because a User failed to organize it within the storage-cleanup period
A problem arising from an error, outage, or policy change of a third-party service (an AI model provider such as OpenAI, or a payment service provider such as Paddle or a PG company)
Damages arising from a User's or a Friend AI's use of a tool
A dispute or damage arising from another person's personal information contained in conversation content or AI-generated output that a User shared externally through a sharing link
A dispute or damage arising from a User modifying, processing, or distributing an exported (downloaded) conversation output

However, this excludes cases involving the Company's intent or gross negligence, and this Article does not limit any liability of the Company that cannot be excluded under the mandatory consumer protection laws of the country in which a User resides.

11. Subscription to and Use of AI Tools

The Company provides AI Tools on a per-tool subscription basis.
A User may use only the AI Tools they have subscribed to, and each AI Tool may include one of the following.

Default monthly usage (number of calls, Call-Based)
A maximum number of concurrent task requests (number of slots, Slot-Based)

Usage is deducted only where an AI Tool call is completed successfully. For a Slot-Based tool, a slot is occupied when a task is requested and is released once the task is completed.
Default monthly usage expires on the next billing date, and additional usage a User has purchased may be used for as long as the subscription is maintained.

※ The concept of monthly usage expiration does not apply to Slot-Based tools.

※ Additional top-up is provided only for call-based tools and does not apply to Slot-Based tools.

The Company may change the composition, provision method, or usage-calculation criteria of an AI Tool according to Service operation or technical needs, provided that where a change materially affects a User's conditions of use, such as a change in the usage method (Call-Based or Slot-Based), the Company will give reasonable advance notice to Users. The procedure for changing subscription terms such as subscription price and provided usage is governed by Article 11-2.
The Company may run an event (such as a free trial) allowing certain Users, such as new sign-ups, to use an AI Tool free of charge for a set period; details such as the target audience, period, content provided, and handling upon conclusion of the event are governed by a separate event policy that the Company sets and announces.

11-1. My Storage Capacity Subscription

The Company may offer a paid subscription that expands the storage limit for RAG documents (the "Storage Capacity Subscription"). A Storage Capacity Subscription is used by selecting one of the capacity options set by the Company; availability, the composition of options, and price are governed by in-service notices and the Pricing Policy.
A change to a higher capacity option (upgrade) applies immediately, and the prorated difference for the remaining period may be charged immediately or added to the next renewal payment.
A change to a lower capacity option (downgrade) and cancellation apply from the next renewal date, and the existing limit is maintained until the renewal date.
Where, as a result of a change in limit (downgrade, cancellation, payment failure, etc.), stored data exceeds the new limit, the Company grants the User a 7-day cleanup period. During the cleanup period, use of stored data (such as reference in AI responses) continues normally, and only new uploads are restricted.
If stored data is not brought within the limit during the cleanup period, the oldest uploaded data is automatically deleted first until the data is within the limit, and deleted data cannot be recovered. The Company provides advance notice to the User both at the start of the cleanup period and before automatic deletion.
If a renewal payment for a Storage Capacity Subscription fails, the cleanup period under paragraph 4, based on the default provided capacity, applies immediately from the time of failure. If payment is restored within the cleanup period, the limit is restored and the stored data remains intact; if payment is not restored within the cleanup period, the subscription terminates after automatic deletion under paragraph 5. Any payment made after termination is treated as a new subscription.
Payment, automatic renewal, and refunds for a Storage Capacity Subscription are governed by Articles 13, 13-1, and 14.

11-2. Changes to Subscription Terms

The Company may change subscription terms, such as subscription price and monthly provided usage, where there is a reasonable basis, such as a change in Service operation or cost structure.
A change disadvantageous to Users will be announced at least 30 days before the effective date through individual notice, such as email, and an in-service notice; the changed terms apply from the first renewal payment occurring after the announced effective date. A change in subscription terms is not applied retroactively to periods already paid for or to additional top-up usage already purchased.
A User who does not agree to a change may cancel their subscription before the renewal; if a renewal payment proceeds, the User is deemed to have agreed to the changed terms.
A change that is not disadvantageous to Users may take effect immediately.

12. Use of Friend AI and Sharing Features

The Company may provide a feature allowing a User's subscribed AI Tool to be used by the User's own AI or by a Friend AI the User has authorized, provided that, depending on the type, features, or operating policy of an AI Tool, Friend AI use of some tools may be restricted or unavailable.
Usage deducted from use of an AI Tool by a Friend AI is calculated against the account of the User who subscribed to that AI Tool.
Where a User permits Friend AI use, RAG reference data belonging to that friend's account may be used in the process of generating an AI response.
In such a case, use of the RAG data is limited to the purpose of generating an AI response in response to the User's request, and occurs only within the scope of the feature provided based on the friend relationship.
A User is responsible for setting and changing a Friend AI's permission to use a tool, and is responsible for the result of a Friend AI's use of a tool or any problem arising therefrom.
The Company does not intervene in the relationship of use between Friend AIs or a dispute between Users, and is not liable for damages arising therefrom.
An AI response and AI-generated output (documents, images, etc.) created in the course of Friend AI use may be viewed outside the Service through a User's sharing link, provided that an original file uploaded by another User, or derived data generated from such a file, is not itself provided or accessible through any route, including a sharing link. In addition, an original file attached during Friend AI use is not stored in any User's storage.
A User may share, through a sharing link, the conversation content and AI-generated output of a chat they created (including 1:1 conversations and 1:N conversations involving a Friend AI) outside the Service, provided that the scope and conditions of conversations that may be shared may be restricted under the Company's Service policy.
A sharing link is valid for 1 day from the time it is created; a third party who accesses it through the link may view the conversation content and AI-generated output but may not modify or edit it. If a User deletes the relevant chat room, the sharing link is immediately invalidated.
Where a conversation in which a Friend AI participated is shared, the Company may, to protect privacy, de-identify the participating AI's profile image and name within the shared view, provided that complete de-identification of personally identifiable information contained in the body of an AI response or in AI-generated output may not be guaranteed due to technical limitations.
Before sharing conversation content and AI-generated output externally, a User is responsible for directly checking whether the content contains another person's personal information or sensitive information. The Company provides notice of the possibility that personal information may be included before sharing, but all legal liability arising from a User's failure to fulfill this obligation to check is attributable to that User.

13. Payment and Refunds

Payment for an AI Tool subscription, purchase of additional usage, and a Storage Capacity Subscription is processed by one of the following methods depending on the payment method and region, and the method applied to each transaction is displayed on the payment screen.

(a) Paddle payments: processed by the Company's payment service provider, Paddle (Paddle.com Market Limited and its affiliates, "Paddle"), as the Merchant of Record. In this case, a User's purchase is also subject to Paddle's Buyer Terms and related policies as notified at the time of payment, which may be reviewed on the payment screen or through a link provided by Paddle.
(b) Payment gateway (PG) payments: the Company acts as the seller, and an electronic payment service provider (hereinafter "PG company") connected through PortOne (포트원), the PG integration platform used by the Company, handles payment processing on the Company's behalf. PortOne is a platform that supports the integration and management of a PG company's payment system, and payment authorization and settlement are performed by each PG company.

The Company does not directly store or process a User's card information or payment method information; such information is processed by Paddle or the relevant PG company in accordance with its own terms and security standards.
The Company confirms only the minimum payment-result information necessary for operating the Service, such as whether payment was completed, subscription status, and the payment identifier.
Refunds are processed in accordance with applicable law and the Company's Refund Policy; Paddle's refund policy applies to Paddle payments, and each PG company's cancellation and refund procedure applies to PG payments, and the timing of a refund may vary depending on the characteristics of the payment method.

※ The Company may require a User to verify their email for the purposes of providing the Service, account security, payment processing, and identifying Users. Where a social login provider has supplied a verified email, the Company may treat it as a verified email; where an email is not provided or its verified status cannot be confirmed, the Company may require additional email verification. If a User does not complete email verification, use of certain Service features (subscription, payment, etc.) may be restricted. (A User is responsible for providing a valid email address they can access and for maintaining its accuracy and accessibility.)

※ For an AI Tool provided on a Slot-Based basis, upon a User's refund, cancellation, or termination of usage rights, any in-progress task, scheduled task, pending request, and temporary data generated in that process for the relevant tool may be deleted. The Company does not guarantee the retention or recovery of such data, and a User is responsible for saving or backing up necessary data in advance where needed.

※ The same applies where a subscription is terminated due to a payment failure. For a Slot-Based tool, any in-progress task, scheduled task, pending request, and related temporary data at the time of termination may be deleted, and the Company does not guarantee their retention or recovery. A User must directly save or back up necessary data before the subscription terminates.

13-1. Automatic Renewal

An AI Tool subscription and a Storage Capacity Subscription automatically renew under the same terms upon expiration of the usage period, provided that where the Company has given notice of a change in subscription terms under Article 11-2, renewals occurring after the announced effective date will renew under the changed terms. For PG payments, recurring payments are made through the payment method (billing key) registered by the User.

The Company, Paddle, or the relevant PG company will inform Users of upcoming charges by reasonable means, such as email or an in-service notice, before the scheduled automatic renewal date.

A User may cancel a subscription from the [Settings > Manage Subscription] menu within the Service at least 1 day before the next billing date.

If a User cancels a subscription, automatic payment will not occur from the next billing date, and the Service may continue to be used for the period already paid for.

14. Summary of the Refund Policy

A User may request a refund only where the relevant AI Tool has not been used at all within 7 days of the payment date.
A refund may be restricted where an AI Tool or usage has been used, even partially. (For a Slot-Based tool, execution of a task request or occupation of a slot is deemed use.)
No refund is provided for a period already paid for upon cancellation of a subscription, and a User may continue to use the Service until the next billing date.
A refund for a payment made through Paddle may be processed in accordance with Paddle's refund policy and Buyer Terms. Where this Article differs from Paddle's policy or the terms notified on the payment screen, the processing standard of Paddle, as the Merchant of Record, prevails, provided that a User's rights under mandatory law are not restricted.
A refund for a payment made through a PG company is processed by the Company in accordance with this Article, the Company's Refund Policy, and applicable law, and the actual refund (payment cancellation) is carried out through that PG company's procedure.
Details concerning refunds are governed by the Refund Policy separately set by the Company.
Given the nature of digital content services, once a User has commenced use of an AI Tool or its usage, withdrawal of an offer to subscribe may be restricted under applicable law (including Article 17(2) of the Act on Consumer Protection in Electronic Commerce for a User residing in the Republic of Korea). A User proceeds with payment after being informed of, and agreeing to, this during the payment process.
A Storage Capacity Subscription applies the expanded storage limit and begins provision of the digital service at the moment payment is completed, so no refund is provided except where the payment is invalid (payment error, duplicate payment, etc.). A change to a lower capacity option (downgrade) is a change in the terms of use that applies from the next renewal date and does not constitute a refund.
This Article does not limit a User's rights (such as the statutory right of withdrawal or right to a refund) under the mandatory consumer protection laws of the country in which the User resides.

15. Limitations of AI Output and Automation Results

AI output and the results of automation features are generated by statistical and probabilistic methods and may be inaccurate, incomplete, biased, or contain hallucinations, depending on data limitations, errors in input information, and characteristics of the model.

The Company does not guarantee fitness for a particular purpose with respect to results arising from these technical characteristics.

16. External Services and Third-Party Liability

The Service may include an API, content, or feature provided by a third party.
A problem arising in connection with the use of a third-party service is governed by that third party's terms and policies, and the Company is not liable for such matters. This includes a response being blocked or restricted under the content policy of an AI language model provider (OpenAI, Anthropic, etc.), and a change in such a policy may affect part of the Service's functionality.

17. Summary of Subscription and Payment

Automatic monthly renewal
Upon a change in subscription terms (price/provided usage): a disadvantageous change is announced 30 days in advance and applies from the next renewal
Subscriptions are provided on a per-tool basis
Monthly provided usage expires on the next billing date
Additional top-up usage expires upon termination of the subscription (provided that, where the Company's refund obligation is recognized under applicable law, that governs)
Storage Capacity Subscription: the expanded limit applies immediately upon payment (a refund is available only for invalid payment); a downgrade or cancellation applies from the next renewal date, and where the new limit is exceeded, automatic deletion follows a 7-day cleanup period
A refund is available if unused within 7 days after payment (processed within 5 to 10 business days)

18. Termination of the Contract

A User may delete their account at any time from the [Settings > Advanced > Delete Account] menu.
Once a request for account deletion is completed, the following will take effect immediately.

(a) Immediate termination of all AI Tool subscriptions and Storage Capacity Subscriptions held by the User
(b) Automatic refund processing for a payment subject to a refund under the Refund Policy
(c) Account logout
(d) Expiration of remaining monthly provided usage and unused additional top-up usage (excluding the portion subject to automatic refund for a payment eligible for a refund)

※ To prevent any subsequent charge to a withdrawn account, the Company immediately deactivates and invalidates the User's active subscription status and automatic-payment linkage on Paddle's side, and the recurring-payment method (billing key) for a PG payment, provided that payment and accounting-related transaction records are retained separately for up to 5 years in accordance with applicable law (see Article 4 of the Privacy Policy).

A 30-day grace period applies from the request for account deletion; if the User logs back in to the same account within that period, the account deletion request is automatically cancelled and the data is restored. Upon expiration of the grace period, all data and usage history are completely destroyed. Details are governed by the Privacy Policy.

※ Even where an account is restored by logging back in within the 30-day grace period, the deactivation and invalidation of active payment status described above cannot be reversed, and a User must re-register a payment method and re-subscribe upon resuming use of the Service.

Where a User violates these Terms, the Company may terminate the account after prior notice, and may terminate it immediately in the case of a material violation.

18-1. Sanctions for Violation of the Terms and Operating Policy

Where a User violates Article 7 (Prohibited Conduct) of these Terms or the Service Operating Policy, the Company may take the following measures, in stages or simultaneously, depending on the severity of the violation.

A prior warning and a request for corrective action
Temporary restriction of use of the relevant feature or AI Tool
Temporary suspension or permanent termination of the account
Expiration of remaining usage (calls/slots) and additional top-up usage
Exclusion of the refund eligibility conditions otherwise applicable under this Policy (denial of a refund)
A claim for damages incurred by the Company or a third party

In the case of a serious and irreparable violation, such as bulk sending of spam, phishing, or fraudulent messages, circumventing sending limits, threatening system stability, or generating a deepfake or sexual synthetic content, the Company may terminate the account and take related action immediately without prior notice.

19. Governing Law and Dispute Resolution

These Terms are governed by the laws of the Republic of Korea, provided that where a User is a consumer residing in a country other than the Republic of Korea, the rights granted to that User under the mandatory consumer protection laws of that country are not restricted by this paragraph.
Both parties will make good-faith efforts to amicably resolve through consultation any dispute arising between the Company and a User.
Where a dispute is not resolved through consultation, a User may apply for dispute mediation to a dispute mediation body such as the Korea Consumer Agency (한국소비자원) or the Consumer Dispute Settlement Commission (소비자분쟁조정위원회) (for a User residing in the Republic of Korea), or to a consumer dispute mediation body in the User's country of residence.
The court having jurisdiction over litigation arising in connection with these Terms or use of the Service is the court determined under the Civil Procedure Act, provided that where mandatory law of the User's country of residence provides for separate jurisdiction, that law governs.

20. License Notice

This Service uses badge resources from Taen and icon resources from Typicons, IcoMoon, Streamline, Flex free icons, and lucide, each of which is subject to the license of its respective author, including CC BY 4.0.

21. Business Information

Information about the business operating this Service is as follows.

Company name: Linkbricks Horizon-AI Inc.
Representative: Yoonsung Ji
Business registration number: 146-87-03100
Mail-order sales registration number: No. 2026-Seoul Seongdong-0928
Address: 7, Yeonmujang 5(o)-gagil, Seongdong-gu, Seoul, Republic of Korea (Hyundai Terrace Tower W1005, Seongsu-dong 2(i)-ga)
Contact: +82 (02) 571-0214
Email: contact@vora.im

Addendum

These Terms take effect on July 21, 2026.