Content Moderation Policy

Applies Globally
Jurisdiction Base: Montenegro (ME)
RYDORA – CONTENT MODERATION POLICY (FORMAL LEGAL VERSION)

This Content Moderation Policy ("Policy") defines the standards, procedures, operational frameworks, technologies, and enforcement mechanisms used by Rydora DOO and Trid Tech DOO ("Rydora", "we", "us", "our") to monitor, analyze, classify, restrict, remove, or otherwise take action on content ("User Content") uploaded, posted, transmitted, or displayed on the Rydora Platform ("Platform").

This Policy forms part of the Terms & Conditions, Community Guidelines, Privacy Policy, and EULA.

Violation of this Policy may result in:

content removal
visibility reduction
temporary restrictions
account suspension
permanent termination
marketplace prohibitions
referral to law enforcement

1. PURPOSE AND SCOPE

1.1. The purpose of this Policy is to:

ensure the safety and integrity of the Platform;
prevent harmful, illegal, or abusive content;
comply with international regulations and industry standards;

establish transparent and fair enforcement mechanisms.

1.2. This Policy applies to ALL User Content including:

photos
videos
comments
messages
Marketplace listings
profile information
metadata
hashtags and captions
AI-generated content

1.3. This Policy applies to every user, regardless of region or local laws, except where stricter laws override.

2. CONTENT CLASSIFICATION FRAMEWORK

Rydora classifies content into enforcement tiers:

Tier 1 – Critical Violations (Immediate Removal + Permanent Ban)

Includes:

child sexual exploitation
human trafficking
terrorism and extremist content
severe violence or gore
distribution of illegal goods
criminal activity facilitation
revenge pornography
credible threats of physical harm
malware or hacking tools

Action:

Instant removal
Permanent account termination
Mandatory evidence retention
Reporting to relevant authorities
Tier 2 – High-Risk Violations (Fast Removal + Temporary or Permanent Restrictions)

Includes:

hate speech
harassment or bullying
sexual content involving adults
depictions of non-graphic violence
significant fraud or impersonation
dangerous acts that encourage harm
self-harm instructions

Action:

Removal within minutes/hours
Account suspension or probation
Mandatory human moderation review
Tier 3 – Medium-Risk Violations (Removal or Visibility Reduction)

Includes:

misinformation with potential harm
borderline explicit content
aggressive or offensive language
misleading Marketplace listings
content violating intellectual property laws

Action:

Content removal or shadow reduction
Warning issued
Repeat offenses escalate to Tier 2
Tier 4 – Low-Risk or Borderline Content (Non-removal Enforcement)

Includes content that is not prohibited but may be:

sensitive
misleading
contextually concerning

Action:

Visibility reduction
Age gating
Limited distribution

3. MODERATION METHODS

Rydora uses a hybrid moderation system consisting of:

3.1. Automated Detection Tools

Rydora uses automated systems such as:

Google Cloud Vision SafeSearch
Rydora internal AI classifiers
computer vision algorithms
text analysis (NLP)
audio analysis
metadata and EXIF analysis
behavioral anomaly detection

These tools detect:

nudity
violence
hate symbols
graphic content
illegal activities
scams and spam
danger patterns

3.2. Human Moderation Teams

Moderators:

review flagged content
make contextual decisions
escalate cases requiring legal intervention
audit automated system outputs
oversee Marketplace compliance

Moderators are trained in:

safety standards
trauma-informed review practices
regional cultural contexts
child safety protection
legal reporting obligations

3.3. User Reporting System

Users may report:

posts
comments
private messages
accounts
Marketplace listings

Reports are automatically logged and triaged.

3.4. Law Enforcement and External Cooperation

Rydora may cooperate with:

police authorities
cybercrime units
anti-trafficking agencies
child safety organizations
regulatory authorities

4. ENFORCEMENT PROCEDURES

4.1. Content Evaluation Process

Content is evaluated through:

automated scanning
contextual analysis
human moderator review
escalation (if necessary)

4.2. Enforcement Actions

Rydora may impose:

content removal
disabling of features
messaging restrictions
comment bans
posting limitations
temporary account suspensions
permanent account deletion
Marketplace bans
IP address blocks
device ID blocks
removal of fraudulent sales earnings

4.3. Emergency Actions

Rydora may immediately:

restrict an account
lock an account
prevent posting
preserve evidence

This applies to:

child exploitation
credible threats
self-harm emergencies
imminent real-world harm

5. MARKETPLACE-SPECIFIC MODERATION

Moderation includes:

verification of listings
detection of counterfeit goods
fraud and scam detection
review of payment anomalies
removal of illegal items
seller reputation tracking
purchase dispute monitoring

High-risk sellers may be suspended without notice.

6. AUTOMATED DECISION-MAKING & USER RIGHTS

6.1. Users acknowledge that decisions may be made using automated systems.

6.2. Users may request:

explanation of enforcement
appeal of removal
review by human moderators

6.3. Automated decisions may be overridden by human moderators when appropriate.

7. EVIDENCE RETENTION AND LOGGING

For safety, compliance, and legal requirements, Rydora may retain:

removed content
moderation logs
chat logs (limited, per legal safety exception)
Marketplace transaction details
device identifiers

Retention timeline:

Tier 1 violations: up to 10 years
Other violations: up to 5 years
Appeals-related data: until resolution

8. APPEALS PROCESS

Users may appeal actions taken against:

content
account
Marketplace listings

Appeals must include:

rationale
contextual explanation
evidence (if applicable)

Rydora will:

acknowledge appeal
review both automated and human decisions
issue final determination

Decisions may be:

upheld
reversed
modified

9. TRANSPARENCY & AUDITABILITY

Rydora maintains:

internal audit logs
decision traceability
regular AI bias testing
transparency summaries (if legally required)

10. MODERATOR CONDUCT STANDARDS

Moderators must:

follow confidentiality rules
maintain impartiality
avoid unauthorized data access
undergo bias and ethics training
adhere to internal safety protocols

11. AI AND MACHINE LEARNING OVERSIGHT

Rydora ensures:

continuous evaluation of AI systems
dataset quality reviews
human-in-the-loop verification
minimization of false positives/negatives
compliance with transparency requirements (EU AI standards)

12. POLICY UPDATES

Rydora may revise this Policy at any time to comply with:

regulatory changes
app store requirements
technological developments
internal operational improvements

Continued use of the Platform constitutes acceptance.

13. CONTACT

For moderation-related inquiries:

📧 support@rydora.me

For appeals:

📧 support@rydora.me

For legal matters:

📧 support@rydora.me