What this app is. What it isn’t. What you could build.
First load takes ~10–30s while we generate the brief. Cached afterward.
One-liner
Canopy is an AI-powered app that filters harmful online content in real time across social media, messaging, and browsing to protect users from cyberbullying, hate speech, and explicit material.
Strengths
Real-time AI filtering of text and images across multiple platforms (Instagram, Twitter, WhatsApp, etc.)
Customizable safety zones with user-defined sensitivity levels for different environments (home, work, school)
Positive reinforcement through 'kindness nudges' that encourage respectful communication
Strong privacy focus: on-device processing minimizes data exposure
Highly rated by parents and educators for protecting teens from online harm
Weaknesses
Frequent false positives: users report legitimate messages being blocked (e.g., 'I'm not okay' flagged as self-harm)
Limited platform support: no integration with TikTok or Discord despite high demand
UI feels outdated; users complain about cluttered settings and poor navigation
No offline mode or local backup for filter rules
Customer support is slow; many reviews mention 'no response after 10 days'
Opportunities
Build a lightweight, open-source version focused on core filtering with transparent model weights
Add granular control per conversation thread (e.g., mute only certain topics in group chats)
Integrate with school/parent portals to allow shared safety policies without full access
Offer a 'filter audit' feature showing why specific content was blocked with explainable AI