Artificial Intelligence (AI) is transforming industries through intelligent automation, predictive analytics, and data-driven decision-making. However, the effectiveness of AI systems relies heavily on the availability of large volumes of data, much of which contains sensitive personal, financial, healthcare, or organizational information. As AI applications continue to expand across cloud computing, healthcare, finance, smart cities, autonomous systems, and critical infrastructure, protecting data privacy while maintaining model utility has become one of the most significant challenges in modern cybersecurity. Privacy-preserving AI seeks to develop techniques that enable AI systems to learn, infer, and collaborate without exposing sensitive information, thereby ensuring compliance with evolving data protection regulations and strengthening public trust in AI technologies.

The International Conference on Cyber Security in Artificial Intelligence (ICCSAI 2027) invites original research contributions on Privacy-Preserving AI, focusing on innovative methods, architectures, and frameworks for securing data, AI models, and user privacy throughout the AI lifecycle. This track provides a platform for researchers, practitioners, industry experts, and policymakers to explore advanced privacy-enhancing technologies, secure machine learning techniques, trustworthy AI frameworks, and privacy-aware governance models. Contributions addressing theoretical advances, practical implementations, privacy attacks and defenses, secure collaborative learning, regulatory compliance, and real-world applications are highly encouraged.

1. Privacy-Enhancing Technologies

• Differential Privacy

• Homomorphic Encryption for AI

• Secure Multi-Party Computation (SMPC)

• Trusted Execution Environments (TEEs)

• Confidential Computing

• Privacy-Enhancing Technologies (PETs)

2. Privacy-Preserving Machine Learning

• Privacy-Preserving Machine Learning (PPML)

• Privacy-Preserving Deep Learning

• Secure AI Model Training

• Privacy-Preserving Model Inference

• Privacy-Aware Model Optimization

• Privacy-Preserving Transfer Learning

3. Federated and Distributed AI

• Federated Learning Security

• Federated Analytics

• Decentralized Machine Learning

• Cross-Silo and Cross-Device Federated Learning

• Collaborative AI with Privacy Guarantees

• Secure Aggregation Protocols

4. AI Privacy Attacks and Defenses

• Membership Inference Attacks

• Model Inversion Attacks

• Model Extraction Attacks

• Data Reconstruction Attacks

• Adversarial Privacy Attacks

• Data Poisoning and Backdoor Attacks

• Privacy Leakage Detection

5. Data Protection and Governance

• Data Anonymization and Pseudonymization

• Secure Data Sharing

• Data Minimization Techniques

• Data Provenance and Integrity

• Secure Data Lifecycle Management

• Privacy Risk Assessment

6. Large Language Models and Generative AI Privacy

• Privacy-Preserving Large Language Models (LLMs)

• Privacy in Generative AI

• Prompt Privacy and Secure Prompt Engineering

• Retrieval-Augmented Generation (RAG) Privacy

• Privacy-Aware AI Agents

• Prevention of Sensitive Information Disclosure

7. Cloud, Edge, and IoT Privacy

• Privacy-Preserving Cloud AI

• Edge AI Privacy

• Internet of Things (IoT) Privacy

• Mobile AI Privacy

• Secure Data Processing in Edge Computing

• Hybrid Cloud Privacy Solutions

8. Explainable and Trustworthy AI

• Explainable AI with Privacy Guarantees

• Trustworthy AI

• Fairness, Accountability, and Transparency

• Privacy-Aware Explainability

• Human-Centered Privacy in AI

• Responsible AI Development

9. AI Governance and Compliance

• Privacy-by-Design for AI Systems

• AI Governance Frameworks

• AI Risk Assessment

• Regulatory Compliance (GDPR, CCPA, HIPAA, EU AI Act)

• Privacy Auditing and Certification

• Ethical AI and Data Protection

10. Secure AI Infrastructure

• Secure MLOps for Privacy-Preserving AI

• AI DevSecOps

• Secure AI Model Deployment

• AI API Privacy and Security

• AI Supply Chain Security

• Continuous Privacy Monitoring

11. Privacy in Domain-Specific AI Applications

• Privacy-Preserving Healthcare AI

• Financial Data Privacy

• Smart City Privacy

• Autonomous Vehicle Data Privacy

• Cybersecurity Data Privacy

• Industrial AI Privacy

12. Emerging Research Directions

• Synthetic Data Generation for Privacy

• Privacy-Preserving Multimodal AI

• Quantum-Resistant Privacy Mechanisms

• Blockchain for Privacy-Preserving AI

• Zero Trust Architectures for Privacy Protection

• Privacy in Autonomous AI Systems

• Secure Multi-Agent AI Collaboration

• Privacy Metrics and Benchmarking