AI Content Guidelines

1. Scope

SynvaxisApp uses generative AI models (from providers such as OpenAI, Anthropic and others) to produce dynamic responses across several surfaces: messaging bots (WhatsApp, Messenger, Telegram, Instagram), the web widget, the admin dashboard, the desktop app and the mobile app. These guidelines apply to any AI-generated output an end user may receive through SynvaxisApp.

2. Prohibited content

We take action on any AI-generated content that falls into any of the following categories:

• Child sexual abuse material (CSAM) or sexualization of minors — absolute prohibition. Such reports are escalated to the NCMEC CyberTipline and competent authorities.
• Self-harm or suicide — instructions, glorification or encouragement. The AI must redirect to crisis hotlines instead.
• Hate speech or discrimination against people based on race, ethnicity, religion, gender, sexual orientation, disability, age or other protected categories.
• Violence, harassment or threats aimed at identifiable people or groups.
• Sexually explicit content directed at adults, when unsolicited or contextually inappropriate.
• Dangerous activities — weapon manufacturing, drug synthesis, explosives or substances that pose real-world safety risks.
• Illegal activity — fraud, evasion, trafficking, money laundering or any conduct typified as a crime in the user's jurisdiction.
• Harmful misinformation — false content about health, elections, emergencies or public safety that may cause concrete harm.
• Privacy violations — disclosure of personally identifiable information (PII) without consent.
• Intellectual property infringement — substantial reproduction of protected works, or trademark impersonation.
• Impersonation or deception — the AI pretending to be a real person, public authority or human when asked directly.

3. How to report

Anyone, whether or not a registered user, may report inappropriate AI content via any of these three paths:

1. Public form: app.synvaxis.com/report-ai-content/en — works without an account, in any browser.
2. From inside the desktop app: system tray icon menu → “Report AI content…”, or the floating “Report AI” button visible at the bottom-right corner of every screen.
3. Direct email: abuse@synvaxis.com for urgent cases or when attachments / evidence are needed.

4. Review process

Every report we receive follows this flow:

1. Automatic acknowledgment — submitting the form generates a ticket number that the reporter can use to follow up by email.
2. Triage (within 2 business days) — Trust & Safety classifies the report by severity. Reports flagged as “immediate harm” are handled the same day.
3. Investigation (within 5 business days for standard reports) — we review conversation logs, the AI model involved and the tenant's prompts to determine whether the output violates these guidelines.
4. Action — we apply one or more of the following:
   • Removal or suppression of the affected content.
   • Hardening of model guardrails (filters, classifiers, system prompts).
   • Suspension or termination of the operator (tenant) account when the root cause was an intentional configuration.
   • Reporting to competent authorities for CSAM or imminent threats.
   • Improvement of public policies if the report exposes a gap in our guidelines.
5. Reporter notification — if the reporter provided an email, we communicate the outcome of the review and the steps taken.

5. Appeals

If you are not satisfied with the decision, you may appeal by replying to the notification email or writing to abuse@synvaxis.com with your ticket number. Appeals are reviewed by a different person from the one who took the initial decision, within a maximum of 10 business days.

6. Emergency cases

If your report involves an immediate threat to a person's life or safety, contact local authorities first (in the U.S.: 911) before filing the report here.

For CSAM, also report directly to NCMEC at report.cybertip.org. That report is independent of ours and is the internationally recognized official channel.

7. Transparency

We publish an aggregated summary of reports received and actions taken in our annual transparency report. Reporter personal data is never published.

8. Contact

Trust & Safety: abuse@synvaxis.com
General support: support@synvaxis.com
Privacy / DPO: privacy@synvaxis.com
Legal entity: Synvaxis LLC, Wyoming, USA