1. Overview
This AI Use Policy describes how Workers, Employers and Worqen itself may use artificial intelligence ("AI") on or in connection with the Worqen Service. It supplements the Terms of Service and the Acceptable Use Policy, and is binding on every User.
We are pro-AI: AI tools can make Workers more productive and Employers better served. We also recognise that AI introduces new risks — deepfakes, misrepresentation, IP infringement, hallucinated outputs, privacy harms — which is why this Policy sets clear rules.
2. Definitions
- AI Tool — any large language model, image, audio, video or multimodal generative model used to assist in or perform tasks on the Service (examples include but are not limited to GPT-class models, Claude, Gemini, Stable Diffusion, Sora, ElevenLabs).
- AI Output — any text, image, audio, video, code or other content generated wholly or in material part by an AI Tool.
- User Content — has the meaning given in the Terms of Service. AI Output you submit to the Service is User Content.
3. Permitted use of AI Tools
You may use AI Tools to:
- Brainstorm, draft, edit, translate, summarise or research, in support of your work on the Service;
- Generate code, copy, design assets, audio, video and other deliverables that you warrant you are entitled to deliver;
- Improve your own profile, listing or pitch — provided the resulting content remains accurate.
You remain fully responsible for any AI Output you submit. Sections 4–6 set out the limits.
4. Prohibited AI uses
You must not, on or via the Service, use AI to:
- Create deepfakes or non-consensual likenesses — generate audio, image, video or text that mimics a real, identifiable person (including by name, voice, face, or other distinctive feature) without that person's verifiable, freely-given consent. This applies regardless of intent and irrespective of whether the content is labelled as AI-generated.
- Misrepresent identity or qualifications — fabricate a profile photo, biographical detail, employer history, education, certification or portfolio piece presented as your own work or your own attributes.
- Generate content unlawful in itself — child sexual abuse material, terrorist content, content promoting self-harm, content that violates EU AI Act Art. 5 prohibited practices (including manipulative subliminal techniques and exploitation of vulnerabilities), or content otherwise prohibited by the Acceptable Use Policy.
- Infringe third-party intellectual property — submit AI Output that you know or should reasonably know infringes copyright, trademark, design, patent or publicity rights of a third party.
- Manipulate the Service — generate fake reviews, fake job listings, fake applications, automated spam or coordinated inauthentic behaviour.
- Train other AI models on the Service or other Users' content — extract data from the Service or from another User's profile, listing or messages to use as training data for any model, except with the explicit written consent of every owner of that data.
- Input a counter-party's confidential materials or another person's personal data into third-party AI Tools — disclose to, or process through, any third-party AI Tool the confidential materials of a Worker or Employer (for example a private brief, unpublished files, source code, or credentials shared with you for an engagement), or personal data about another person, unless you have a lawful basis or that person's consent to do so and doing so is consistent with your own obligations as a data controller under applicable data-protection law. This complements the rule against sharing content that contains personal data of others without a lawful basis in the Acceptable Use Policy.
- Bypass identity verification — use AI-generated documents, faces or voices to attempt to defeat the KYC flow described in the Identity Verification Policy.
5. Disclosure obligations
5.1 Disclose AI use on request
If a counter-party (Worker or Employer) asks whether and how AI Tools were used to produce a deliverable or to draft a profile or listing, you must answer truthfully. Refusal to answer such a request is a breach of this Policy.
5.2 Disclose AI use proactively where it materially matters
Where AI Output forms a material part of a deliverable in a category in which it would be material to the counter-party (for example, ghostwriting copy, voiceover work, code substantially produced by an AI assistant, image-generation work), you should disclose AI involvement up-front, before contracting. Failing to do so may amount to misrepresentation and expose you to a dispute outcome against you.
5.3 Disclose synthetic media to viewers (EU AI Act Art. 50)
Deliverables that are or contain "deep fakes" or other AI-generated content depicting real people, places or events must be marked as artificially generated or manipulated, in a manner that is clear and discernible at latest at the time of first interaction or exposure.
6. Intellectual property in AI Output
- You represent and warrant that you have the right to submit each piece of AI Output as User Content under the Terms of Service, and that doing so does not infringe a third party's rights.
- You are responsible for complying with the licence terms of every AI Tool you use, including any commercial-use, attribution or training-data restrictions.
- The licence you grant Worqen over User Content (Section 9 of the Terms of Service) applies to AI Output the same as to any other User Content.
7. Selling AI-as-a-Service on Worqen
If your offering on the Service consists of AI services (for example, prompt engineering, AI-content generation, AI-powered analysis), you must additionally:
- Identify the underlying AI Tool(s) and any model versions where this is reasonably knowable;
- Make any limitations of the AI Output (hallucinations, training-data cut-off, accuracy variability) plain in your listing description;
- Comply with the EU AI Act's transparency obligations applicable to general- purpose AI deployers, and refrain from offering AI services that fall within Art. 5 prohibited practices;
- Comply with any sectoral rules (medical, legal, financial advice) that apply irrespective of whether AI is in the loop.
8. How Worqen uses AI
8.1 No training on your content (today)
Worqen does not currently use User Content to train, fine-tune or evaluate any first-party machine-learning model, and does not make User Content available to third parties to do so.
8.2 Bilateral opt-in if that ever changes
If we ever introduce a feature that uses User Content from a contract to train models, the affected Worker and the affected Employer will both be required to give a separate, affirmative, freely-revocable opt-in. Either party may opt-out for that contract; opt-out will exclude the contract from any training data set. We will notify Users at least30 days before such a feature is enabled.
8.3 Operational AI use
We may use AI tools internally to detect spam, abuse, fraud, sanctioned activity, and to triage or summarise dispute evidence. AI is decision-support, not the final decider: adverse decisions affecting your account are reviewed by a human Worqen team-member before they take effect, consistent with EU GDPR Article 22.
9. Enforcement
Breach of this Policy is treated as a breach of the Acceptable Use Policy and is enforceable through the same remedies — content removal, account suspension or termination, refusal to release escrow, dispute outcomes against the breaching party, and cooperation with law-enforcement where required.
10. Contact
Questions: legal@worqen.com. Reports of AI-related abuse: abuse@worqen.com.