š± I Built ESCAPE: Touch Grass to Unlock Social Media ā Powered by Offline AI
Description and use
ESCAPE replaces a socialāmedia password with an outdoor quest, using onādevice AI to generate missions and verify nature photos, thereby reducing screen time.
Problem
Users spend excessive time scrolling on social media, leading to reduced outdoor activity and wellābeing.
Outcome
ESCAPE demonstrates that an offline AIāpowered app can encourage outdoor activity and reduce screen time while still providing socialāmedia access.
Workflow
- Move: step outside and complete a walking goal using phoneās step sensor and timer.
- Discover: find an interesting natural element such as grass, a tree, flowers, or water reflection.
- Capture: photograph the discovery, preview the image, and tap Analyze. ESCAPE checks the evidence locally.
- Complete the quest to earn a temporary socialāmedia unlock.
Evidence
ESCAPE encourages you to get outside, walk, touch grass, and rediscover nature.
ESCAPE checks the evidence locally.
I tested image analysis with my phone disconnected from the internet, and ESCAPE still approved the nature photograph!
ESCAPE is an open-source Android prototype built using Flutter and Kotlin.
Gemma 3 1B runs locally through LiteRT-LM, generating creative outdoor missions without requiring continuous internet connectivity.
- Record date: 2026-10-10T21:56:21.186Z