kodykebab/adscience

Adscience is an extremely early-stage (4 days old) prototype likely created for a hackathon or as a proof-of-concept. While the combination of Fully Homomorphic Encryption (FHE) and blockchain for ad targeting is conceptually interesting and technically challenging, the project currently lacks any markers of defensibility. With 0 stars and a fresh codebase, it functions primarily as a demonstration of FHE primitives (likely leveraging Zama's library or similar FHE-compatible chains like Fhenix or Inco). The technical moat is low because the heavy lifting is done by underlying FHE libraries; once those libraries are mature, the 'matching' logic becomes a standard implementation. The competitive landscape includes established privacy-centric ad networks like Brave (BAT), which have significant user bases and actual browser integration. Furthermore, while frontier labs (OpenAI/Google) are not building on-chain FHE ad networks, they are heavily invested in 'Privacy Sandbox' and differential privacy techniques for ads, which solve the same business problem with much lower computational overhead than FHE. The 'on-chain' constraint limits scalability, making it highly susceptible to displacement by more efficient off-chain ZK-proof or MPC-based privacy solutions within a short timeframe.