Terms of Service - Smartegy.ai

Terms of Service

Smartegy.ai

Effective: February 1, 2026

1. Introduction and Subject of the Agreement

1.1. Provider Information

These Terms of Service (hereinafter: 'Terms') govern the use of the Smartegy.ai service (hereinafter: 'Service').

  • Developer and rights holder: Chain Advisory Kft.
  • Registered office: 1037 Budapest, Táborhegyi út 18/f, Hungary
  • Company registration: 01-09-326168
  • Tax number: 26361518-2-41

Joint controller and operator:

  • Erba 96 Kft.
  • Registered office: 1142 Budapest, Stubnyai utca 4., Hungary
  • Company registration: 01-09-561901
  • Tax number: 12175590-2-42
The Service may be sold through reseller partners. Reseller partners act as independent data controllers in their own sales processes.

1.2. Subject of the Service

Smartegy.ai is an artificial intelligence-based business intelligence (BI) platform that enables Users to communicate with their business data in natural language, create automated analyses, reports and visualizations, and automate business processes.

The central element of the platform is an AI orchestrator that interprets the User's natural language requests and engages specialized Agents to execute tasks. Agents perform dedicated, controlled tasks such as: executing database queries (with read-only access); generating Excel, PDF and other format reports; creating interactive charts and map visualizations; sending emails and SMS; web search and market intelligence gathering; competitor and company analysis; file management and processing (with OCR); executing scheduled, recurring tasks; and AI-powered image and video generation. Agent operations are governed by the agentic workflow architecture detailed in Section 8.8.

The Service is designed for B2B (business) use. Each User receives their own isolated workspace where data and configurations are stored and processed separately from other users.

pages.terms.sections.intro.acceptance.title

pages.terms.sections.intro.acceptance.content

2. Definitions

The terms used in these Terms have the following meanings:

  • Provider: Chain Advisory Kft. (developer, rights holder) and Erba 96 Kft. (joint controller and operator), the operators of the Smartegy.ai platform.
  • User: A natural or legal person who registers and uses the Service.
  • Service: The Smartegy.ai platform and all its features.
  • Business data: Corporate/business information uploaded by the User to the platform or derived from databases connected to the platform.
  • AI analysis: Data analysis and response generation performed by artificial intelligence.
  • Subscription: A contract for a specific period entitling the use of the Service.
  • Workspace: An isolated environment provided for the User where data storage and processing takes place.

3. Service Description

3.1. Core Features

The Smartegy.ai platform provides the following main features:

  • Natural language queries: Ask questions and receive answers in Hungarian, English, or German without programming skills.
  • AI-powered data analysis: Automatic pattern recognition, trend identification, and business insight generation.
  • Automated reporting: Create customizable reports in Excel and PDF formats with interactive charts and map visualizations.
  • Multi-format export: Export data and analyses in Excel, PDF, and CSV formats.
  • Data integration: Connect to external data sources and databases.
  • Specialized Agents: The platform uses a modular Agent system where each Agent handles a specific task domain (e.g., database querying, report generation, email sending, web search, competitor analysis, scheduled tasks, image and video generation, etc.). The available Agents depend on the subscription tier and company permission settings.
  • Scheduled tasks: Users can set up recurring, timed tasks (e.g., daily reports, weekly summaries) that the system executes automatically in the background.
  • Communication features: Send emails and SMS directly from the platform with automatic recipient resolution.

pages.terms.sections.serviceDescription.multiTenant.title

pages.terms.sections.serviceDescription.multiTenant.content

4. Registration and Account Management

4.1. Registration Requirements

Registration is required to use the Service. The User must be at least 18 years old, provide real and accurate information, and have a valid email address.

4.2. Account Security

The User is obligated to choose a secure password, keep it confidential, and immediately report any unauthorized access to the account to the Provider.

4.3. Account Suspension

The Provider may suspend or delete the account if the User violates the Terms, poses a security risk, or fails to fulfill payment obligations.

5. Subscriptions and Payment

5.1. Subscription Models

The Service is available on a subscription basis. Final end-user prices are always determined individually, taking into account the requested features, the volume of data, the method of data storage, and the level of anonymization if required by the client. The Provider always prepares a custom quote, preceded by assessments and consulting discussions.

5.2. Payment Terms

Subscription fees are payable in advance on a monthly or annual billing cycle. Prices do not include VAT unless otherwise stated.

5.3. Price Changes

The Provider reserves the right to modify prices. Users will be notified of price changes at least 30 days in advance.

5.4. Late Payment

In case of late payment, the Provider may restrict features after 15 days and suspend access after 30 days.

5.5. Refund Policy

For annual subscriptions, a full refund may be requested within the first 14 days of the contract (right of withdrawal). After this period, subscription fees are non-refundable.

5.6. Pilot (Trial) Period

The Provider may offer a pilot (trial) period to the User based on an individual agreement for testing the Service in a live environment. The pilot period duration is a minimum of 1 week and a maximum of 1 month, as specified in the individual agreement.

During the pilot period, a usage limit (rate limit) is enforced: the User may send a maximum number of requests (prompts) per day to the AI system as specified in the individual agreement. This value is typically between 1 and 50 according to the individual agreement. Upon special request, the Provider may increase this limit during the pilot period.

During the pilot period, the User has access to set up scheduled (timed) tasks. The User assumes full responsibility for scheduled tasks and their results. Scheduled tasks are an integral part of the pilot period and are automatically deactivated upon pilot expiration, unless the User enters into a permanent contract.

During the pilot period, all provisions of these Terms apply to the User in full, including ethical rules (6.3.), prohibited activities, data protection provisions, liability rules, and all other contractual obligations. The pilot period does not exempt the User from contractual obligations.
During the pilot period, the Smartegy.ai platform does NOT perform anonymization on data by default. The custom anonymization option detailed in Section 8.6 is available exclusively under a permanent contract. During the pilot period, the AI system works directly with the original data.
The operator of Smartegy.ai reserves the right to supervise user activity on the https://smartegy.ai platform during the pilot period. The pilot period may only be initiated under a written agreement (including agreements concluded via electronic correspondence). The purpose of supervision is to ensure service quality and prevent irregular usage. The legal basis for supervision is the legitimate interest in contract performance (GDPR Article 6(1)(f)). If the Client objects to this, please notify the Data Protection Officer (DPO) at [email protected].

Upon expiration of the pilot period, the User may decide to enter into a permanent subscription. If no permanent contract is concluded within 5 business days following the expiration of the pilot period, the Provider will suspend access. Data and reports generated during the pilot period will be available for export for 30 days following the suspension.

6. User Rights and Obligations

6.1. User Rights

The User is entitled to use the Service features, upload and analyze data, use the AI assistant, and export or delete their data.

6.2. User Obligations

The User is obligated to use the Service only for lawful purposes, comply with all applicable laws, respect third-party rights, and provide accurate information.

6.3. Prohibited Activities

It is prohibited to circumvent security systems, upload malicious code, access other users' data without authorization, overload the Service, upload illegal content, or resell the Service without permission.

Ethical use obligation:

The User is obligated to use the Service exclusively for ethical purposes. It is expressly PROHIBITED to use the Service for the following purposes, even if the User bypasses the system's built-in ethical safeguards through any technical method, trick, or prompt manipulation:

  • Generating pornographic, sexually explicit, or obscene content
  • Creating war propaganda, weapons-related guides, or military strategies
  • Any form of incitement to hatred, discrimination, or harassment (racial, religious, gender-based, sexual orientation-based, etc.)
  • Promoting or facilitating terrorism, violence, or self-harm
  • Unauthorized collection, processing, or disclosure of personal data (doxxing)
  • Intentional creation of disinformation, fake news, or misleading content
  • Development of malicious software, viruses, or cyberattack tools
  • Supporting fraud, money laundering, or any illegal financial activity
  • Infringing copyrights, trademarks, or intellectual property rights
  • Automated screening, ranking, or pre-selection of resumes (CVs)
  • AI-based evaluation, scoring, or suitability assessment of job candidates
  • Automated employee performance evaluation, behavioral analysis, or HR decision-making based on personal data.

In case of violation of ethical rules, the Provider is entitled to immediately and unilaterally terminate the contract, immediately suspend the User's access, delete the affected content, and notify the authorities if necessary. A User whose contract has been terminated due to ethical violations is not entitled to a refund.

6.4. Responsibility for Uploaded Data

The User is fully responsible for the legality and accuracy of uploaded data and for having the necessary rights to process the data on the platform. Detailed rules regarding uploaded data are set out in Section 6.6.

6.5. Resale and White-Label Prohibition

Reselling the Service under one's own brand name (white-label), sublicensing to third parties, or mass automated resale of Service outputs is expressly PROHIBITED without the Provider's prior written consent. Reseller partners may only sell according to the commercial model approved by the Provider.

6.6. Responsibility for Uploaded Data

Smartegy.ai assumes no responsibility for the content of data uploaded by the Client or its users, neither during the pilot period nor thereafter. This applies to data submitted via file upload, the attachment function, or any other channel. The uploading party bears sole responsibility for the content of uploaded data.

Only business-related data may be uploaded to the system. The ethical rules set out in Section 6.3 also apply to uploaded content. Prohibited content includes in particular:

  • Pornographic or sexually explicit content
  • Hate speech, incitement, or racist content
  • Content glorifying or encouraging violence
  • Data related to illegal activities
  • Unauthorized processing of personal data (GDPR violation)
  • Content infringing intellectual property rights
  • Any content involving the exploitation of minors
  • Content supporting or promoting terrorism
  • False, misleading, or deceptive content intended to cause business harm

Uploading prohibited content may result in immediate suspension of service and extraordinary termination of the contract, including the right to claim damages.

7. Data Ownership and Management

7.1. Business Data Ownership

The User remains the exclusive owner of business data uploaded or connected to the platform. The Provider does not acquire any ownership rights over this data.

7.2. Data Usage Restrictions

The Provider uses the User's business data exclusively for providing the Service, preparing analyses, and resolving technical issues.

7.3. Data Protection

User data is not sold or shared with third parties for marketing purposes. Business data is NOT used for AI model training. All data is stored encrypted within the EU.

7.4. Data Export and Deletion

The User may export their data at any time. Upon contract termination, 30 days are available for data export, after which the data will be permanently deleted.

Important note on audit logs: The audit logs detailed in Section 8.11 will be retained after a deletion request or contract termination until the end of the 90-day retention period, if it has not yet expired following the deletion request. This ensures the possibility of regulatory compliance and retrospective investigation of security incidents. Audit logs are automatically and irreversibly deleted after the retention period expires.

8. AI Services and Limitations

8.1. AI Operation

The Service uses artificial intelligence technologies (Anthropic Claude models via the AWS Bedrock platform) for data analysis and response generation. AI processing takes place exclusively within the EU (AWS eu-central-1, Frankfurt).

8.2. Accuracy Limitations and AI Hallucinations

AI-generated responses and analyses are informational in nature. AI responses are not guaranteed to be 100% accurate. It is the User's responsibility to verify AI responses before making decisions.

Due to the inherent nature of the AI system, it may occasionally generate incorrect, inaccurate, incomplete, or non-existent information (so-called 'AI hallucination'). This is a known limitation of LLM technology, not a software bug. The Provider accepts NO liability for any direct or indirect damages, lost profits, or business losses arising from AI hallucinations, erroneous responses, or inaccurate analyses.

The User is obligated to independently verify every AI-generated response, analysis, calculation, and recommendation from independent sources before making business decisions based on them, forwarding them to third parties, or using them in any official document.

8.3. Does Not Replace Professional Advice

The AI service does NOT replace professional financial, legal, tax advice, auditing services, or any qualified professional's opinion.

8.4. Prohibition of Automated Decision-Making

Under Article 22 of the GDPR, it is PROHIBITED to use the Service for solely automated decision-making that produces legal effects concerning natural persons or similarly significantly affects them. AI responses may only be used as decision-support tools; the final decision must always be made by a human. This includes in particular: hiring, dismissal, or evaluation of employees; screening, ranking, or pre-selection of resumes (CVs); evaluation, scoring, or suitability assessment of job candidates; employee performance evaluation or behavioral analysis; credit assessment or financial decisions about natural persons; healthcare decisions; any decision that affects the rights or freedoms of natural persons.

8.5. User Responsibility

The User acknowledges that they are solely responsible for business decisions made based on AI responses. The Provider accepts no liability for consequences of decisions made based on AI responses.

8.6. Data Processing and Anonymization

Important note: Business data is NOT anonymized by default during AI processing, as masking and hiding personal data from the LLM requires custom, client-specific solutions due to the inherent nature of LLM technology. Data protection is ensured through isolation, encryption, EU-only processing, and the prohibition of model training.

Custom anonymization option:

If the client insists on data anonymization, it can be implemented based on a jointly developed plan. The process is as follows: (1) We discuss with the client the scope and details of the data to be anonymized. (2) We develop a detailed anonymization plan, which may include the use of pseudonyms, masking techniques, or other data protection solutions. (3) The plan defines the process: who does what - i.e., what modifications are needed on the data source side and how we can assist with the implementation. (4) The plan describes in detail the functional limitations that anonymization will impose on usage, and what special usage practices must be followed (e.g., using pseudonyms in queries, special identifier systems, etc.). It is important to note that anonymization always entails certain usage restrictions, which the client will be informed of when accepting the plan.

8.7. Database Access and Data Protection

It is important to clarify that the AI system (LLM) does NOT have direct access to the client's database. The LLM only sees the database structure (schema) - i.e., table names, columns, and data types - as well as the user's request formulated in natural language. Database queries are generated by a specialized agent in SQL format, but query execution takes place exclusively in the locally hosted or the client's own database, not in the LLM. The LLM therefore never sees the entire database - it only knows the structure and the user's query parameters, then receives the query results for formulating the response.

8.8. Agent-based (Agentic) Workflow Architecture

Smartegy.ai employs an agent-based (agentic) architecture that ensures secure and deterministic data processing as follows:

USER→ORCHESTRATOR (LLM)→AGENT→LOCAL DB / SERVICE
←←← response ←←←
1The user asks their question in natural language.
2The orchestrator (central LLM) interprets the request and decides which specialized agent to involve (e.g., database query agent, report generator, email sender, etc.).
3The specialized agent executes the task deterministically: SQL queries run in the local database, file generation occurs on the local server, email sending goes through a dedicated service.
4The result is returned to the orchestrator, which formulates the response for the user.
Why is this solution deterministic? The LLM serves exclusively as a reasoning and decision-making component: it decides WHAT needs to be done and HOW to formulate the response. The actual execution (data queries, file generation, email sending) always takes place in local, isolated code that the LLM cannot bypass. The agents have strictly regulated permissions (e.g., the database query agent can only execute SELECT statements, meaning it can only read data, not modify it). This separation ensures that data never leaves the protected environment, and operations always run within controlled boundaries.

8.9. Third-Party System Connections and Liability

The Service allows Users to connect their own databases and external systems. The User bears sole responsibility for: (a) providing and verifying the legality of access credentials required for connected systems; (b) ensuring that the handling of data stored in connected data sources complies with applicable data protection regulations; (c) the secure management of access credentials (passwords, API keys).

The Provider accesses connected external systems exclusively with the access level expressly authorized by the User. For database connections, the system operates with read-only (SELECT) privileges and does not modify data. The User may revoke access at any time.
The Provider is not liable for the availability, data content, data quality, or security deficiencies of connected external systems. The outage or malfunction of external systems does not constitute a breach of contract by the Provider.

8.10. Special Categories of Data (GDPR Article 9)

The Service is NOT suitable for and NOT authorized for the processing of special categories of personal data under Article 9 of the GDPR. It is expressly PROHIBITED to upload to the platform or process through connected databases the following data categories: data revealing racial or ethnic origin; data concerning political opinions, religious or philosophical beliefs; trade union membership data; genetic and biometric data for the purpose of uniquely identifying natural persons; health data; data concerning a natural person's sex life or sexual orientation.

If the User's business data necessarily contains special categories of data (e.g., in the case of a healthcare provider), the User must notify this in writing before entering into the contract. In such cases, a specific Data Processing Agreement (DPA) and additional technical protective measures are required, which are subject to individual agreements.
If the User uploads special categories of data to the system in violation of the above prohibition, the Provider assumes no liability for any resulting data protection incidents, regulatory fines, or third-party claims.

8.11. Audit Logging and Data Access

The Provider maintains detailed audit logs of all user interactions, which include: all user queries (prompts), responses generated by the AI system, the identifiers and operations of agents used, database queries and their results, generated files and exports, as well as timestamps and user identifiers.

The purpose of audit logging is to ensure the secure operation of the Service, monitor compliance with ethical rules, troubleshoot issues, and protect User rights.
The contractual partner with an end-user agreement (the data owner) - whether directly or through a reseller partner - may request the release of audit log data in database dump format in writing. The request must be sent to the [email protected] email address. The Provider will fulfill the request within 10 business days.
Audit logs are retained for 90 days. In the event of contract termination, audit logs are also available during the 30-day period provided for data export. Audit logs are automatically deleted after the 90-day retention period expires, unless legislation requires longer retention.

8.12. Fair Use and Capacity

The Service operates on a fair use principle. The Provider reserves the right to temporarily restrict usage if a User's activity results in disproportionately high resource consumption that jeopardizes the service quality of other users. In such cases, the Provider will notify the User in advance and offer a solution (e.g., a higher-tier subscription). Custom capacity requirements are subject to individual agreements.

8.13. Data Breach Management and Notification

The Provider acts in accordance with Articles 33-34 of the GDPR in the event of a personal data breach.

The Provider will notify affected Users and the competent supervisory authority (National Authority for Data Protection and Freedom of Information, NAIH) without undue delay and no later than 72 hours after becoming aware of the data breach. If notification within 72 hours is not possible, the Provider will provide reasons for the delay.

The notification will include: the nature and approximate scope of the incident; the categories and approximate quantity of data affected; the possible consequences of the incident; the measures taken or planned by the Provider to mitigate the damage; the contact details of the Data Protection Officer (DPO).
Upon becoming aware of the incident, the Provider will immediately take the necessary technical and organizational measures to prevent further data leakage, minimize damage, and restore system integrity.

9. Intellectual Property

9.1. Provider's Intellectual Property

The Smartegy.ai platform, including software, user interface, algorithms, documentation, and trademarks, is the exclusive intellectual property of Chain Advisory Kft.

9.2. Limited License

The Provider grants the User a non-exclusive, non-transferable, revocable license to use the Service for the duration of the subscription.

9.3. Prohibited Activities

It is prohibited to reverse engineer, modify, copy, distribute the Service, or use trademarks without permission.

10. Service Availability

10.1. Availability and SLA

The Provider strives for continuous Service availability. Target availability: 99.5% monthly, excluding scheduled maintenance and force majeure events.

Service Level Agreement (SLA): If the monthly availability falls below the target of 99.5% (excluding scheduled maintenance and force majeure events), the User is entitled to the following compensation: between 99.0%-99.4%: credit equivalent to 5% of the monthly subscription fee; between 95.0%-98.9%: credit equivalent to 15% of the monthly subscription fee; below 95.0%: credit equivalent to 30% of the monthly subscription fee. The compensation claim must be submitted in writing by the User to the [email protected] email address within 30 days following the affected month. The credit will be applied in the next billing cycle. The maximum amount of compensation shall not exceed 30% of the monthly subscription fee. If the monthly availability falls below 95.0% for 3 consecutive months, the User is entitled to extraordinary termination of the contract with immediate effect and a pro-rata refund of the subscription fee for the remaining subscription period.

10.2. Scheduled Maintenance

Users will be notified of scheduled maintenance at least 48 hours in advance. Maintenance is scheduled for low-traffic periods when possible.

10.3. Force Majeure

The Provider is not liable for service outages caused by events beyond its control. Force majeure events include in particular: natural disasters (earthquakes, floods, pandemics); war, acts of terrorism, civil unrest; government measures, changes in legislation; cyber attacks (DDoS, ransomware); outages of the AI provider (AWS Bedrock) or service disruptions of the AWS eu-central-1 region; internet infrastructure failures; and other events beyond the Provider's control. In case of an AWS Bedrock service outage, the Provider will promptly notify Users and strive to provide an alternative solution as soon as possible.

10.4. Data Backup & Recovery

Smartegy.ai performs automatic daily backups of every customer instance, database, audit log, and stored data. Backups are stored geo-redundantly within the EU (Hetzner, Germany).

The previous day's state can be restored from backups. The Provider assumes no liability for data loss occurring between backup cycles.
Upon client request — except during the pilot period — a custom backup strategy may be developed. This may include more frequent backups or storing backups on infrastructure provided by the client. Custom backup terms are always subject to individual agreement.

11. Limitation of Liability

11.1. Service Provided 'As Is'

The Service is provided on an 'as is' basis. The Provider makes no warranty for uninterrupted operation, accuracy of AI responses, or suitability for all purposes.

11.2. Exclusion of Liability

The Provider is not liable for indirect or consequential damages, lost profits, data loss, or consequences of decisions made based on AI responses. This exclusion of liability does not apply to damages resulting from the Provider's intentional or grossly negligent conduct.

11.3. Limitation of Liability Amount

The Provider's maximum liability shall not exceed the amount actually paid by the User in the 12 months preceding the damage event.

11.4. User's Indemnification Obligation

The User shall indemnify the Provider against any claims arising from Terms violations, uploaded content, or unlawful activities.

12. Confidentiality

Both parties are obligated to treat business secrets, technical information, and other confidential data received from the other party as confidential. The confidentiality obligation remains in effect for 3 years after contract termination.

13. Contract Duration and Termination

13.1. Contract Duration

The contract is effective from registration. Monthly subscriptions are for an indefinite period, annual subscriptions are for one year, and automatically renew.

13.2. Regular Termination

The User may cancel their subscription at any time effective at the end of the next billing period. For annual subscriptions, notice must be given at least 30 days before renewal.

13.3. Immediate Termination

The Provider may terminate immediately if the User seriously violates the Terms, engages in illegal activity, or fails to pay for more than 30 days.

13.4. Post-Termination Data Handling

Upon contract termination, 30 days are available for data export. After this, data will be permanently deleted, unless legislation requires longer retention.

14. Modifications to Terms

The Provider reserves the right to modify the Terms. For material changes, at least 30 days' advance notice will be provided. Continued use of the Service constitutes acceptance of the modified terms.

15. Governing Law and Dispute Resolution

15.1. Governing Law

These Terms and the legal relationships arising from them are governed by Hungarian law.

15.2. Dispute Resolution

The parties shall primarily resolve disputes amicably. If this fails, the parties agree to the exclusive jurisdiction of the Budapest Metropolitan Court or, depending on the value in dispute, the Buda Central District Court.

15.3. Online Dispute Resolution

The European Commission's online dispute resolution platform is available at: https://ec.europa.eu/consumers/odr. Note: Given that the Service is primarily intended for B2B (business) use (see Section 1.2), the ODR platform is only relevant if the User qualifies as a consumer under applicable legislation.

16. Miscellaneous Provisions

16.1. Severability

If any provision of the Terms is invalid, it does not affect the validity of other provisions.

16.2. No Waiver

If the Provider does not enforce any right, it does not constitute a waiver of that right.

16.3. Entire Agreement

These Terms together with the Privacy Policy constitute the entire agreement between the parties regarding the Service.

16.4. Notices

Contract-related notices may be validly sent via email or through the platform's notification system.

17. Contact

For questions, comments, or complaints, please contact us:

Chain Advisory Kft. (developer, rights holder)

  • Registered office: 1037 Budapest, Táborhegyi út 18/f, Hungary

Erba 96 Kft. (joint controller and operator)

  • Registered office: 1142 Budapest, Stubnyai utca 4., Hungary

Customer support is available on business days from 9:00 AM to 5:00 PM CET.

We typically respond to emails within 24-48 hours.