Usage Policy

1. Purpose and Scope

(1) This Usage Policy (the “Policy”) describes how all users of the Sleak Platform must use it safely, lawfully and responsibly. Any person who accesses the Platform agrees to this Policy.

(2) The Policy forms part of the agreement with Sleak. It supplements the General Terms of Use (Allgemeine Nutzungsbedingungen), the DPA, the Documentation and any notices in the Platform.

(3) The order of precedence set out in the Agreement (Section 1 of the General Terms of Use) applies. For usage, security and AI governance matters, this Policy prevails over the Documentation and general product information. The German version is legally binding.

2. Core Principle: AI Supports People

(1) Sleak supports training, coaching, feedback and assessments but does not replace human judgment. Sleak develops and operates the Platform taking into account the requirements of the EU AI Act (Regulation (EU) 2024/1689) applicable to Sleak as provider, as and when they become applicable to Sleak; this constitutes neither a guarantee nor an agreed quality (Beschaffenheitsvereinbarung).

(2) AI Outputs are probabilistic and may contain errors, omissions or bias. Review Outputs before relying on them, especially in sensitive contexts.

(3) AI Outputs must not be the sole or decisive basis for decisions about people; a qualified person always makes the decision.

3. User Responsibilities

All users must:

• use Sleak only within the Agreement, Documentation and applicable law.
• carefully review Inputs and Outputs and not use Outputs that are obviously incorrect, inappropriate or unlawful.
• protect access credentials and promptly report unauthorized access.
• not upload data, recordings or content they are not authorized to process.
• not enter special categories of personal data or any unnecessary personal data (data minimization).
• report violations, security issues or concerning Outputs to the customer administrator or to support@sleak.ai.

Customers, owners and administrators are additionally responsible for the lawful use of the Platform under the General Terms of Use, the Documentation and applicable law.

4. Prohibited Uses

The following uses are prohibited; they include in particular the practices prohibited under Art. 5 of the AI Act:

• using Sleak as the sole or decisive basis for legally significant or similarly material decisions about people.
• designing or using processes that bypass or merely simulate meaningful human review.
• using Sleak to infer, predict, assess or proxy protected characteristics (including ethnic origin, gender, sexual orientation, religion, belief, disability, health, age, trade union membership, national origin).
• using content or criteria that directly or indirectly discriminate.
• emotion recognition (prohibited in the workplace and education under Art. 5 of the AI Act): using Sleak to infer emotional states from voice, face, body or similar signals, or combining it with external tools for that purpose.
• using Sleak for covert, disproportionate or unlawful surveillance.
• using Sleak for social scoring, manipulative or exploitative techniques or other practices prohibited under Art. 5 of the AI Act.

5. Data Protection, Recordings and Sensitive Data

(1) Users must not upload or process recordings, transcripts, documents or personal data unless a legal basis, authorization or required notice exists.

(2) Special categories of personal data must not be entered into or processed through the Platform. This includes in particular health data, genetic data, biometric data for unique identification, and data revealing ethnic origin, political opinions, religion or belief, trade union membership, sex life or sexual orientation. Government identification numbers, payment data and credentials must likewise not be entered. Users limit their inputs to the data necessary for the agreed purpose (data minimization); Sleak may technically restrict the entry of such data or remove it.

(3) Data of persons under sixteen (16) years of age must not be processed through Sleak. For persons aged sixteen (16) or seventeen (17), processing is permitted only where a valid legal basis and appropriate safeguards exist.

(4) Recordings, transcripts and analysis results may be made accessible only to persons who need them for a permitted purpose.

6. Platform Integrity and Security

The following actions are prohibited:

• reverse engineering, decompilation, disassembly or examination of source code, models, algorithms or protection mechanisms, except where mandatorily permitted by law.
• bypassing access controls, usage limits, security measures, suspensions, throttling or technical safeguards.
• using adversarial inputs, prompt injection, jailbreaks or similar techniques to bypass safety, moderation or product limits.
• uploading or transmitting malware, malicious code, exploits or content that endangers systems, data or users.
• unauthorized scans, penetration tests, load tests, scraping, benchmarking or automated use without Sleak's prior consent.
• creating multiple accounts to circumvent usage limits, checks, security measures or commercial restrictions.

7. Content, Deception and Harmful Use

Users must not use Sleak to create, upload, distribute or support content that is unlawful, fraudulent, defamatory, discriminatory, harassing, threatening, pornographic, exploitative, self-harm-promoting, violence-glorifying, extremist, terrorist, harmful to minors or otherwise harmful.

Users must also not:

• imitate real persons without their consent or misrepresent their identity.
• present AI-generated content as human-generated where disclosure is legally or contextually required.
• deceive people about whether they are interacting with an AI system.
• use Sleak for fraud, phishing, social engineering, scams, spam or targeted disinformation.
• enter third-party confidential information or trade secrets without authorization.

8. Unauthorized Commercial Use

Without an express agreement with Sleak, it is prohibited to:

• sell, rent, lease, transfer, sublicense or use access to the Platform as a service bureau for third parties, unless this was expressly allowed in writing in the form of a special agreement.
• use Sleak or Outputs to develop, train or improve competing products, AI systems, models or datasets.
• use Sleak trademarks, logos or product elements without permission.
• use usage data, Outputs or product information in a way that violates the Agreement, third-party rights or Sleak's legitimate interests.

9. Enforcement, Appeals and Reporting

(1) Sleak may remove content, restrict features, throttle use, suspend accounts or suspend access where there are concrete indications of a breach of this Policy, the Agreement or applicable law or where a security, integrity or availability risk exists.

(2) Immediate action remains permitted where necessary to avoid harm, legal violations, security risks or impacts on other customers.

(3) Appeals: If a user or customer considers an enforcement measure to be incorrect, they may object within thirty (30) days to legal@sleak.ai, stating the measure, the reasons and any supporting information. Sleak reviews appeals in good faith and responds within a reasonable period. The measure generally remains in effect during the review.

(4) Reporting and protection of reporters: Violations, security issues or concerns may be reported to support@sleak.ai or privacy@sleak.ai; reports may be made anonymously to the extent permitted by law. Sleak protects good-faith reporters against retaliation. A customer's staff may additionally raise concerns internally or through competent authorities (e.g. data protection or labor authorities).

(5) Material changes to this Policy will be notified in accordance with the Agreement.