Skip to main content

🪞 Impersonation Detection

Automatically identify and remove impersonators pretending to be admins or trusted figures, ensuring your community stays safe without the need for manual monitoring. This feature uses Fren One's custom AI model to protect your members from fraud and social engineering attacks.

Where to configure it

In the webapp: Open your workspace, select the Telegram group under Channels in the sidebar, then select AI Features, then the AI Moderator tab. Impersonation detection is the Impersonation rule under New members. The webapp has its on/off switch, which saves instantly; the action taken on detection is chosen in the bot menu.

In Telegram (bot menu): the /settings menu route below still works.

Overview​

Fren One's Impersonation Detection leverages a custom AI model deployed by Fren One specifically designed to detect profile impersonation in Telegram communities.

What is Impersonation Detection?​

In online communities, impersonators often:

  • Pose as admins or moderators: Copy names, photos, and usernames
  • Offer fake assistance: Pretend to provide support to steal information
  • Share malicious links or malware: Trick users into clicking dangerous links
  • Scam members with fraudulent support claims: Request passwords, private keys, or personal data
  • Create fake urgency: "Your account will be banned unless you..."
  • Request sensitive information: Ask for wallet seeds, API keys, or login credentials

Manually spotting these threats is time-consuming and error-prone. Fren One automates this process using AI, scanning new and existing members for suspicious behavior or impersonation attempts.

Common Impersonation Tactics

Scammers often change just one letter in a name (e.g., "Admin" → "Adm1n"), use similar profile pictures, or copy admin bios. Fren One's AI is trained to detect these subtle differences.

How It Works​

Fren One's Impersonation Detection works by:

  1. Keeping a Trusted Baseline: the bot compares members against your admins' profile information (name, username, profile picture). Admin profile pictures are refreshed automatically when they're out of date; there's no manual update step
  2. Comparing New Members: Every new user joining is compared against the admin baseline
  3. Monitoring Profile Changes: When existing members change their profile information, the bot checks for impersonation
  4. AI Analysis: The AI model analyzes visual similarity (profile pictures), text similarity (names/usernames), and behavioral patterns
  5. Automatic Action: When impersonation is detected, the bot applies your chosen action: Mute member, Kick member, Ban member (the default) or Alert admins
  6. Admin Notification: Optionally notifies admins of detected impersonators

The AI uses advanced computer vision for profile picture comparison and natural language processing for name/username similarity detection.

Setup Guide​

Step 1: Enable Impersonation Detection​

  1. Use the /settings command in your chat
  2. Navigate to 🤖 AI → 🪞 Impersonation Detection
  3. Toggle the feature 🟢 Enabled
Enable Early

Turn on Impersonation Detection as soon as you add Fren One to your community to establish a baseline before impersonators appear.

Step 2: Admin baseline​

There's no manual step. The bot compares members against your current administrators' names, usernames and profile pictures, and refreshes admin profile pictures automatically when they're out of date, so admin changes are picked up on their own.

Step 3: Real-Time Monitoring​

From now on, all new users joining your group and existing users who change their profile information will be automatically scanned for impersonation attempts.

Automatic scanning triggers:

  • User joins the group
  • User changes display name
  • User changes profile picture
  • User changes username
What happens on detection

The bot applies the action chosen in the bot menu: mute, kick, ban (the default) or alert admins. Detections are reported to your admins so you can review them.

Troubleshooting

Legitimate users being flagged?

  • Check if their profile genuinely resembles an admin
  • Ask user to change profile to be less similar
  • Reverse the action manually (unmute or unban) if it was a false positive
  • Consider switching the action to alert admins while you review

Impersonators getting through?

  • Verify Impersonation Detection is Enabled (🟢)
  • Ensure the bot has the permissions your chosen action needs (restrict or ban members)

False positives after admin changes?

  • Admin profile pictures refresh automatically, but a brief window after an admin changes their profile is possible
  • If admins plan profile changes, you can switch the rule off during the change and back on afterwards

Scan not working?

  • Verify bot has admin privileges
  • Check bot has permission to see member list
  • Ensure the bot can restrict or ban users

How to review banned impersonators?

  • Go to Telegram group settings
  • Navigate to "Restricted Users"
  • Review recently banned accounts
  • Unban if false positive

On this page