How Flamingo’s AI Moderation Works for the Social Feed

Overview

The Flamingo app's social feed allows residents to post public messages that are visible to their neighbors. While most posts contribute positively to community interaction, some may be inappropriate or misdirected—such as complaints, profanity, or maintenance requests.

To maintain a respectful and relevant social environment, Flamingo uses AI-powered moderation to review posts before publication.

Feature Functionality

1. 🔄 Post Submission & Review Process

Step-by-Step

Step 1: Resident submits a post to the social feed using the Flamingo resident app.
Step 2: AI reviews the post for:

• Inappropriate language (e.g., profanity, slurs)

• Complaints better suited for the complaint ticket system

• Maintenance requests better suited for the maintenance request system

• Violations of community or building-specific policies

• Spam, solicitation, or inflammatory content

Step 3: AI takes action based on building settings:

Buildings with Post Approval OFF

• If appropriate, the post is published immediately.

• If inappropriate, the resident receives a rejection message (see below).

  • The resident can then choose to:

    • Edit the post (which is re-evaluated by AI)

    • Submit for manual approval by staff

• If submitted for manual approval, AI-rejected posts are automatically approved after a predefined approval time limit if not manually addressed.

Buildings with Post Approval ON

• If appropriate, the post is sent to staff for manual review (standard approval flow).

• If inappropriate, the same rejection and re-submission flow as above applies.

Comment Moderation

AI moderation applies not only to posts, but also to comments added to any post.

AI Moderation Prompt

Here is the standard prompt Flamingo uses to guide the AI system when evaluating user-submitted content:

AI Moderation Prompt

You are the content moderator for a resident app. Residents can publicly post text and images visible to their neighbors and management. Your job is to ensure all posts align with community standards and building-specific policies.

Evaluate the following post and determine whether it is appropriate for public display. Flag for:

• Inappropriate language (profanity, slurs, discriminatory terms)

• Misplaced complaints (should go to complaint ticketing)

• Maintenance issues (should go to maintenance ticketing)

• Spam, disrespectful tone, inflammatory language

• Inappropriate or explicit images (nudity, violence, graphic content)

• Promotion of illegal or prohibited activities

• Building-specific policy violations
  [INSERT BUILDING SPECIFIC SOCIAL FEED POLICIES HERE]

🔸 Note: Unless otherwise specified in building policies, solicitation and commercial promotion are allowed.
🔸 Posts seeking help (e.g., missing packages) are permitted if phrased as community inquiries, not accusatory messages toward management.

Here is the post to evaluate:
[INSERT POST CONTENT HERE]

If the post is acceptable, return only: APPROPRIATE
If the post is not acceptable, return only one of these rejection reasons:

• Complaint Misplacement

• Maintenance Request Misplacement

• Inappropriate Language

• Violation of Community Standards

• Violation of Building Specific Policies

• Inflammatory Content

• Inappropriate Image Content

• Other

Rejection Reasons & Resident Messages

If the AI flags a post, the resident receives one of the following messages and options to edit or submit for manual review.

Additional Notes

• If your building has a “Social Feed Policy” content tile in the Digital Resident Guide, Flamingo will automatically pull in those building-specific policies to guide AI evaluation.

• All AI-rejected posts submitted for approval are auto-approved after a predefined time if not manually reviewed.

• Residents can always edit and resubmit posts.