A discussion on novel methods for hijacking AI agents through social engineering, the evolution of DDoS attacks, the legacy of Zero Trust, and the glaring security flaws in AI training data apps.
A novel framework called zk-promises is introduced, enabling stateful anonymous credentials with Turing-complete state machines and asynchronous callbacks. This allows for robust moderation, such as banning or reputation updates, for anonymous users without compromising their privacy, using zk-objects and zero-knowledge proofs to ensure state integrity.
An exploration of encrypted computation, detailing how techniques like homomorphic encryption and multi-party computation can enable machine learning on encrypted data. The summary covers the core mathematical principles, real-world use cases, and open-source libraries to build more private and trustworthy AI systems.
A technical deep dive into Apple's Private Cloud Compute (PCC), exploring its novel architecture for running sensitive AI workloads with cryptographic privacy guarantees. The talk covers the core requirements, key components like remote attestation and transparency logs, and how these concepts can be applied by developers today.