Relying on cloud computing, big data, and Internet of Things (IoT), artificial intelligence (AI) has developed rapidly and been widely researched in various fields such as machine learning, recommender system, and natural language processing. In recent years, remarkable surges in AI have led to several innovations including autonomous vehicles and connected IoT devices in our homes. AI is even contributing to the development of a brain-controlled robotic arm that can help a paralyzed person feel again through complex direct human-brain interfaces.

However, security issues have posed serious challenges for the widespread application of these technologies. Without the security technology, AI may not only bring convenience, but also disaster. Cryptography is the core technology to solve security problems, how to adapt it to AI is a key issue. The state-of-the-art mainly focuses on secure multiparty computation, homomorphic encryption, secure outsourcing computation, and federated learning. In addition, verifiable technology has also become important to ensure correctness and integrity of AI systems. However, there is still a lack of appropriate cryptographic technology that can balance security and accuracy for AI. On the other hand, some cryptographic primitives having been used in AI are high consuming in computation or communication, which greatly impacts the usability and practicability. Exploring lightweight cryptographic alternatives and new promising cryptographic techniques for AI is still a challenging research direction.

This Special Issue welcomes original research and review articles on the topic of cryptographic schemes and protocols for artificial intelligence.

Potential topics include but are not limited to the following:

• Security and privacy risks/models for AI system
• Security and privacy issues for AI in blockchains
• Security and privacy issues for AI in IoT
• Security and privacy issues for AI in specific application scenarios
• Security and privacy issues for distributed AI systems
• Cryptographic schemes and protocols for AI
• Secure multiparty computation protocols for AI
• Homomorphic encryption scheme for AI
• Secure outsourcing scheme for AI
• Privacy-preserving data mining and machine learning
• Privacy-preserving federated learning
• Privacy-preserving natural language processing
• Privacy-preserving graphics