AUTHORS

Call for Papers

We invite original research on generative AI security, adversarial ML, privacy, and trustworthy systems. All accepted papers will be published in IEEE proceedings and indexed in IEEE Xplore.

Submit your paper Submission guidelines →

Research Tracks

Topics & Categories

GAISS 2026 features six focused tracks. Select the track that best fits your work when submitting.

Track 01

LLM Security & Robustness

Defense mechanisms for large language models against adversarial attacks, prompt injection, jailbreaking, and model manipulation.

Prompt injection detection & defense
Jailbreak prevention techniques
Model hardening & robustness evaluation
Adversarial text generation
Safety alignment methods

Track 02

Privacy-Preserving AI

Techniques for training and deploying AI while protecting sensitive data, including federated learning, differential privacy, and secure inference.

Federated learning security
Differential privacy for neural networks
Secure multi-party computation
Privacy-preserving inference
Data anonymization for ML

Track 03

AI for Cybersecurity

Using generative AI and ML for threat detection, incident response, vulnerability analysis, and autonomous security operations.

AI-powered threat detection
Automated incident response
Vulnerability discovery with LLMs
Malware analysis & classification
Security operations automation

Track 04

Adversarial Machine Learning

Attacks and defenses across the ML pipeline, including data poisoning, model extraction, membership inference, and evasion attacks.

Data poisoning & backdoor attacks
Model extraction & stealing
Membership inference attacks
Evasion & transferability attacks
Certified robustness methods

Track 05

Secure AI Development

Engineering practices for building secure AI systems, including MLOps security, supply chain integrity, and secure-by-design frameworks.

Secure ML pipelines & MLOps
Model supply chain integrity
AI bill of materials (AI-BOM)
Secure model serving & deployment
Reproducibility & auditability

Track 06

Ethics, Alignment & Governance

Responsible AI practices including bias mitigation, transparency, fairness, regulatory compliance, and AI governance frameworks.

Bias detection & mitigation
AI transparency & explainability
Regulatory compliance (EU AI Act, etc.)
AI governance frameworks
Human-AI alignment research

Call for Papers

Submit Your Research

Submissions undergo double-blind peer review. Accepted papers appear in IEEE proceedings and are indexed in IEEE Xplore. We invite full papers, short papers, and workshop proposals on generative AI, ML security, privacy-preserving computation, and closely related topics.

Representative themes include adversarial robustness, prompt and tool-use abuse, federated and distributed learning security, model extraction and misuse, and safety evaluation. See the guidelines for formatting and page limits.

Submission Guidelines