Find 2 Earn

Core mechanic: Players receive a stylized image depicting a real-world location. Using digital maps, they identify the place and submit coordinates or a canonical map link. First valid submission wins.

Why this works:

AI-resistant (2-4 year horizon):

• Maps are artistically redrawn, removing literal features

• Cultural clues require context (local landmarks, architectural styles, historical references)

• No reverse image search databases exist for stylized cartography

• Multimodal AI (GPT-Vision class) currently lacks training data for this task

Skill-based:

• Pattern recognition (identifying terrain, road layouts, coastlines)

• Geographic knowledge (narrowing regions by climate, infrastructure, vegetation)

• Research ability (cross-referencing clues with historical data)

• Speed (first correct answer wins)

Naturally social:

• Players form teams to pool knowledge.

• Discord/X chats share partial progress; Hunt Log provides public ordering and proof.

• Fog Hunt lets players buy “negative info” (hide wrong sectors) to focus the search.

• Public Hunt Log provides transparent proof of outcomes