Moderation Policy -

Version: 1.0
Date of Last Update: [TO BE FILLED — publication date]
Effective Date: [TO BE FILLED — publication date]

1. About This Policy

1.1 This Moderation Policy explains how Collab Media Group Ltd ("Collabable", "we", "us", "our") identifies, reviews, and acts on Content and conduct on the Platform at https://collabable.ai that breaches our Terms of Service, our Acceptable Use Policy, or the law.

1.2 This document is published for transparency. It is written for both our Users and our internal staff, and it supports our duties under the Online Safety Act 2023 (OSA), the Electronic Commerce (EC Directive) Regulations 2002, the Defamation Act 2013, the Digital Markets, Competition and Consumers Act 2024 (DMCCA), and related UK law.

1.3 This Moderation Policy should be read together with:

our Terms of Service (clause 8 — Content moderation; clause 9 — Notice and action; clause 10 — Statement of reasons and complaints);
our Acceptable Use Policy (rules on what may and may not be posted); and
our Privacy Policy (how we handle personal data in moderation).

1.4 Defined terms have the same meaning as in our Terms of Service.

2. Our Moderation Principles

2.1 Our moderation is guided by the following principles:

Proportionality. We apply the least intrusive measure that addresses the issue effectively.
Consistency. We apply the same rules to all Users, regardless of Subscription plan, verification status, or visibility.
Fairness. We give Users an opportunity to be heard before taking significant action against their Account, unless doing so would be impracticable or would compromise an investigation or the safety of others.
Respect for fundamental rights. We balance enforcement with Users' rights to freedom of expression, privacy, and non-discrimination.
Transparency. We tell Users when and why their Content is restricted, except where the law prevents us from doing so.
Human oversight. Final decisions on significant moderation actions are taken by trained human reviewers, not by automated tools alone.
Continuous improvement. We review this policy and our moderation practices at least once per year, and whenever a new risk or duty arises.

3. What We Moderate

3.1 We moderate all types of Content and conduct on the Platform, including:

public profiles (display name, biography, photographs, media samples, topics, links);
Studio listings (title, description, photographs, pricing, availability);
search and discovery metadata (categories, keywords, tags);
reviews and ratings;
private messages between Users;
collaboration requests;
Account-level behaviour (patterns of logins, rate of messaging, interactions with other Users);
third-party content that we import (for example, reviews from a connected Google Business Profile).

3.2 We do not routinely monitor the contents of private messages. We access private messages only when we are alerted to a potential breach (for example, through a User report or an automated flag) and only to the extent necessary to assess and respond to that alert.

4. Triggers for Moderation

4.1 A moderation review can be triggered by:

a User report submitted by email to contact@collabable.ai;
an automated signal from the tools described in clause 5;
a request from a trusted flagger (a person or organisation with recognised expertise in identifying illegal Content — in the United Kingdom, we give priority to reports from bodies such as the Internet Watch Foundation for child sexual abuse material);
a request from a law enforcement agency, court, or regulator (including the Information Commissioner's Office, Ofcom, the Competition and Markets Authority, or the Financial Conduct Authority);
a formal legal notice (for example, a copyright or defamation takedown request);
our own proactive review, for example where a recurring pattern of abuse has been identified and we decide to investigate further.

5. Automated Tools We Use

5.1 We use a limited set of automated tools to help us spot Content and conduct that may breach our rules. These tools do not take the final decision on any significant moderation action. Their role is to surface items for human review.

5.2 The tools we currently use include:

Keyword and pattern filters applied to profiles, listings, messages, and reviews to detect common categories of abuse (for example, obvious slurs, scam templates, contact-detail exfiltration, phishing signatures).
Image classification on uploaded photographs to flag content that appears to be sexually explicit, violent, or otherwise prohibited under clause 3 of the Acceptable Use Policy.
Rate-limiting and anomaly detection on sign-in attempts, password resets, messaging volume, and collaboration requests, to identify brute-force attacks, credential stuffing, or spam campaigns.
Duplicate-content detection to identify repeated identical messages (spam) and duplicate Studio listings.
Fake-review detection that flags reviews whose pattern (timing, wording, account age, review source) departs significantly from the norm for verified reviewers on the Platform.
Trust Engine signals, which combine verification data, activity history, and reliability indicators to raise or lower the priority of a review item.

5.3 What the automated tools do not do. They do not suspend, terminate, or permanently restrict any Account on their own. They do not set pricing, ranking, or access to features based on profiling that produces legal or similarly significant effects on you. See clause 6 of our Privacy Policy for further details.

5.4 We test, audit, and tune these tools regularly to reduce false positives, avoid bias, and ensure they work as intended. We document changes to automated tooling in our internal moderation log (see clause 12).

6. Human Review

6.1 All significant moderation actions — including Content removal beyond obvious spam, visibility reductions, feature restrictions, Account suspensions, and Account terminations — are reviewed by a member of our moderation team before they take effect, except where immediate action is required to protect Users or to comply with a binding legal obligation.

6.2 Our moderation team is:

trained in the rules in our Terms of Service, Acceptable Use Policy, and this Moderation Policy;
trained in the categories of illegal Content under the Online Safety Act 2023, including how to recognise terrorism content, child sexual abuse material, and threats to life that must be escalated immediately;
trained in the fundamental rights framework (freedom of expression, privacy, non-discrimination) that applies to moderation decisions;
supported with access to legal advice for difficult cases;
supported with appropriate wellbeing measures (rotation, time limits on exposure to distressing material, access to confidential counselling), in line with the Online Safety Act 2023 duty to have regard to the safety of persons exercising moderation functions.

6.3 Escalation. Reviewers escalate unclear, novel, or high-impact cases to a senior moderator or to our legal advisers. Cases involving potential criminal offences, imminent risk to life, or matters of public safety are escalated immediately and may be reported to the relevant authorities in accordance with clause 10.

7. Categories of Moderation Action

7.1 We choose from the following actions, applying the least intrusive measure that addresses the issue effectively:

Action	When it applies	Reversible?
No action	The item does not breach our rules. The report is closed.	—
Warning	Minor first-time breach. The User is notified and asked to correct the issue.	Yes
Content correction	We ask the User to edit Content that is factually inaccurate or misleading.	Yes
Content removal	Specific Content is deleted or hidden.	Yes (we can restore if the decision is reversed)
Visibility reduction	Content remains visible to the author but is demoted in search results and recommendations, or hidden from public listings.	Yes
Feature restriction	The User temporarily loses access to specific features (for example, sending new requests, posting reviews, listing Studios), while keeping their Account.	Yes
Account suspension	The Account is disabled for a defined period. The User cannot sign in or use the Platform during that time.	Yes
Account termination	The Account is permanently closed.	Rarely reversible
Legal action	We pursue civil remedies against a User who has caused serious harm to us or another User.	—
Referral to authorities	We refer suspected criminal Content or conduct to the police or another competent authority, in accordance with clause 10.	—

7.2 Aggravating and mitigating factors. When choosing an action, our reviewers consider:

the severity and impact of the breach;
whether the breach was deliberate or inadvertent;
the User's history of prior breaches;
the age of the Account and its verification status;
the vulnerability of any affected party;
any corrective steps the User has already taken;
the User's response to a warning or request for correction.

8. Priority Categories

8.1 Certain categories of Content or conduct are always treated as the highest priority and reviewed immediately upon detection or report:

Child sexual abuse material (CSAM) — reported to the National Crime Agency and, where appropriate, to the Internet Watch Foundation, and the Content removed without delay;
Terrorism content — reviewed, removed, and, where appropriate, reported to the Counter Terrorism Internet Referral Unit;
Credible threats of violence or imminent risk to life — escalated to the police without delay;
Intimate image abuse — removed without delay and the reporter offered information on the Revenge Porn Helpline and other support resources;
Human trafficking, modern slavery, or exploitation indicators — escalated to specialist authorities;
Fraud or financial crime affecting Users — escalated to Action Fraud where appropriate, with the User informed.

8.2 For these priority categories, we may act before contacting the affected User, where contacting the User first would prejudice the investigation or the safety of a person.

9. Timeframes

9.1 We aim to handle moderation matters within the following indicative timeframes:

Event	Target timeframe
Acknowledge a User report	Within 48 hours of receipt
Decide on a report of potentially illegal Content	Without undue delay, and normally within 7 days
Decide on a report of other rule breaches	Without undue delay, and normally within 14 days
Act on a priority category (clause 8)	Without delay upon confirmation
Respond to an internal complaint against a moderation decision	Within one month of receipt
Respond to a lawful request from a public authority	Within the time specified in the request or, failing that, without undue delay

9.2 These are targets, not guarantees. Complex cases, investigations, or matters that require legal advice may take longer; we inform the affected User if this is the case.

10. Cooperation with Authorities

10.1 We cooperate with the police, the National Crime Agency, Ofcom, the Information Commissioner's Office, and other competent authorities in the United Kingdom when they make lawful requests for information or action.

10.2 We disclose personal data to authorities only when we are legally required to do so, when we have a lawful basis under the UK GDPR, and after a careful review of the request. Where possible, we ask the requester to use the appropriate legal process (for example, a court order under Schedule 7 of the Investigatory Powers Act 2016 or a notice under Schedule 2 of the Data Protection Act 2018).

10.3 We document every disclosure, including the requester, the legal basis, the scope of the data disclosed, and the decision-maker on our side. This record is retained for 5 years.

10.4 Where we proactively report suspected criminal Content or conduct to an authority (for example, a report to the police regarding a credible threat to life), we do so in accordance with UK law and our Privacy Policy.

11. Statement of Reasons and Right to Complain

11.1 When we take a moderation action against you or your Content, we notify you as soon as reasonably practicable and include:

what action we have taken (for example, removal, visibility reduction, suspension);
the territorial scope and duration of the action, where applicable;
a description of the Content or conduct concerned;
the rule or legal ground we relied on;
whether automated means were used to identify or decide on the matter;
the remedies available to you, including our internal complaint system.

We do not provide a statement of reasons where we are prohibited from doing so by law, where doing so would compromise an active investigation, or where the matter concerns clear spam or manifestly illegal Content and the User has been previously warned.

11.2 Internal complaint. You can contest any moderation decision free of charge through the internal complaint system described in clause 10.2 of our Terms of Service, for a period of six (6) months from the decision. Complaints are reviewed by a member of our team who was not involved in the original decision, and are not decided solely on the basis of automated means.

11.3 Outcome of a complaint. If we conclude that the original decision was wrong, we reinstate the Content or Account, notify you, and update our internal records to prevent similar errors in future. If we conclude that the original decision was correct, we confirm it and explain our reasoning.

11.4 External dispute resolution. If you are a Consumer and you are not satisfied with the outcome of our internal complaint-handling, you may be entitled to refer the matter to an alternative dispute resolution body in accordance with the Alternative Dispute Resolution for Consumer Disputes (Competent Authorities and Information) Regulations 2015.

11.5 Legal remedies. Nothing in this Moderation Policy restricts your right to bring proceedings in a competent court or to raise a concern with a regulator or law enforcement agency.

12. Record-Keeping, Review, and Transparency

12.1 We keep internal records of all moderation decisions, including:

the nature of the trigger (report, automated flag, authority request, proactive review);
the Content or conduct concerned;
the action taken and the reason;
whether automated tools contributed to the decision and how;
the identity of the reviewer;
any subsequent internal complaint and its outcome.

12.2 We retain these records for 5 years from the date of the decision, in accordance with our Privacy Policy, to enable us to identify patterns, improve our tools, and respond to any subsequent investigation or complaint.

12.3 Annual review. We review this Moderation Policy and our moderation practices at least once per year, and more often if significant changes occur in the law, in regulator guidance (including Ofcom codes of practice issued under the Online Safety Act 2023), or in the risks facing our Users.

12.4 Transparency. We will publish aggregate information about our moderation activity at least once per year, including the volume of reports received, the volume of actions taken by category, and the outcome of internal complaints, subject to our obligations to protect personal data and active investigations.

13. Safeguarding our Moderation Team

13.1 We recognise that moderation staff may be exposed to distressing Content. We comply with our duties under the Online Safety Act 2023 to have regard to the safety and wellbeing of persons performing moderation functions. Our measures include:

rotation of reviewers on distressing categories;
time limits on continuous exposure to high-severity material;
access to confidential mental-health support;
clear escalation paths for cases a reviewer is not comfortable handling;
ongoing training and debriefs.

14. Changes to This Policy

14.1 We may update this Moderation Policy from time to time to reflect changes to the Platform, applicable law, Ofcom codes of practice, or our internal processes. The "Date of Last Update" at the top of this document indicates when the most recent changes took effect.

14.2 Where the changes are material, we will notify Users by email and/or through a prominent notice on the Platform at least 15 days before the changes take effect.

15. Contact Us

For any question about this Moderation Policy, or to make a report:

Collab Media Group Ltd
20-22 Wenlock Road, London, N1 7GU, United Kingdom
Email: contact@collabable.ai