The article discusses the limitations of decentralized artificial intelligence (DeAI) due to a lack of diverse and secure data. While centralized AI models have access to vast amounts of data, DeAI models struggle to find precise and varied datasets to train on. However, cryptography, specifically zero-knowledge proofs like zero-knowledge fully homomorphic encryption (zkFHE) and zero-knowledge TLS (zkTLS), can offer a solution. zkFHE enables computations on encrypted data, allowing DeAI models to learn from privacy-protected datasets. zkTLS extends this principle to internet communication, allowing users to prove possession of certain data without revealing it. By leveraging these cryptographic techniques, DeAI can access and utilize data from web2 while maintaining privacy and decentralization. The implementation of zkFHE and zkTLS is computationally intensive and requires standardization and interoperability for widespread adoption. However, the potential rewards include a more democratic and equitable AI future.



Other News from Today