ZkSciAi
A meme marketplace where a chain of AI agents generates scientific memes, and another chain of agents automatically ZK verifies them.
Demo
About
The problem it solves My ZK-powered AI meme generation platform tackles key issues in education, AI ethics, privacy, and decentralization, offering innovative solutions to entrenched problems. 1. Simplification of Complex Epistemic Content The platform leverages AI to distill esoteric knowledge into succinct, visual memes, rendering abstruse concepts comprehensible. By transcending traditional didactic methods, it facilitates cognitive accessibility for diverse learners, offering a heuristic and humorous approach to complex material. 2.Verification through Zero-Knowledge Proofs (ZKP) ZKPs ensure the veracity of AI-generated memes without revealing input data, securing both content accuracy and user privacy. This cryptographic validation establishes trust in AI outputs, preventing misinformation and enhancing epistemic integrity. 3.Preservation of Privacy and Data Sovereignty ZKPs obfuscate sensitive data, maintaining privacy while ensuring the authenticity of content. Users are assured that their inputs are not disclosed or exploited, preserving confidentiality in a decentralized system. 4. Democratization of Educational Resources By allowing on-demand meme generation, the platform bypasses traditional, elitist educational barriers, offering personalized learning experiences. Memes cater to visual, cognitive learners, enhancing engagement and knowledge retention. 5.Active User Engagement and Incentivization The platform fosters participatory content creation, wherein users contribute educational memes, incentivized through a blockchain-based tipping system. This meritocratic structure motivates high-quality contributions and promotes continuous learning. 6.Ethical AI Transparency ZKPs ensure the ethical operation of AI algorithms, eliminating opacity and biases. This fosters a transparent AI environment where content generation is verified and accountability is paramount.
How it's made
Challenges we ran into It was my first time using verify, so initially i faced many problems implementing it. It was facing lots of errors while compiling my circom, sometime r1cs was not generating, sometimes wasm was not generating etc. I faced lots of issues while integrating these stuffs in js for proof generation. Implementing Chain of Agents was more difficult than i thought It was very difficult to interact with edu chain, thats why front end was not as good as we expected.