1st Workshop on
AI for Fraud and Abuse 

at KDD 2026

Jeju, Korea on August 9–13, 2026 

AI for Fraud and Abuse: Fraud and abuse are no longer isolated domain problems but are increasingly enabled by automated agents and cross-platform generative AI frameworks. AI-FAB 2026 explores our ability to bridge domain-specific silos to identify universal patterns and scalable AI-driven defenses. Crucially, this workshop seeks to complement model performance research with the economic and quantifiable realities of fraud; interests aim to move beyond simple detection to model attacker ROI, address fundamental survivorship bias in labeling, and explore the optimization of friction through mechanism design. We invite submissions at the intersection of a comprehensive range of AI methodologies related to adversarial ML, industry specific challenges for defenses, and their practical applications related to fraud detection, quantification, and robustness. 

The workshop covers a comprehensive range of AI methodologies, industry specific challenges, and their practical applications that fall into the below core technical topics.

● Adversarial and Generative AI Challenges

○ Generative AI Threat Landscape

■ Detection of synthetic identities, deepfakes, and AI-generated phishing

■ Mitigating Large Language Model (LLM) exploitation for automated social engineering

■ Synthetic data generation using GANs for defense-side data augmentation

○ Adversarial Robustness and Red Teaming

■ Defending against evasion, poisoning, and model inversion attacks

■ Automated adversarial testing (Red Teaming) of high-stakes AI systems

○ Adaptive & Autonomous Systems

■ Reinforcement learning and agentic frameworks that enable self-evolving security strategies

■ Decomposition of complex fraud investigations into manageable autonomous tasks

■ Real-time adaptation to rapid "concept drift" in adversarial behavior

● Core AI and Data Mining Methodologies

○ Anomaly Detection and Representation Learning

■ Unsupervised and self-supervised learning for rare event detection

■ Deep learning architectures for high-dimensional, imbalanced datasets

■ Cost-sensitive classification and adaptive thresholding

○ Graph Analytics and Network Science

■ Graph Neural Networks and Graph Foundation Models (GNNs and GFMs) for identifying fraudulent bad actor rings

■ Community detection for coordinated inauthentic behavior and botnets

■ Temporal graphs for modeling evolving adversarial relationships

○ Multimodal Intelligence

■ Integrating LLMs, NLP, and Computer Vision for cross-channel detection

■ Multimodal fusion for analyzing transactions alongside communication logs and behavioral metadata

● Holistic Application Domains (Theory to Practice)

○ Economic and Financial Integrity

■ Banking, Fintech, and Digital Payment fraud (conventionally referred to as 1st party and 3rd party fraud)

■ Cryptocurrency forensics and Blockchain-based crime detection

■ Anti-Money Laundering (AML) and Know Your Customer (KYC) innovations

○ Digital Platform and Infrastructure Abuse

■ E-commerce abuse (e.g., return fraud, promo abuse) and Ad fraud (Invalid Traffic, Content Safety Violations, etc)

■ Cloud infrastructure exploitation and resource exhaustion attacks

■ Systemic manipulation of search, recommendation, and ranking algorithms

○ Social and Behavioral Integrity (Trust & Safety)

■ Coordinated influence campaigns and digital dis/misinformation

■ Online toxicity, harassment, and behavioral platform abuse

■ Detection of high-harm patterns (e.g., child trafficking, illicit sales)

● Operational, Ethical and Regulatory Frameworks

○ Explainable & Collaborative AI

■ Interpretable models for regulatory compliance (e.g., AML/KYC) and human-in-the-loop decision support

■ Federated learning approaches that enable privacy-preserving fraud detection across multiple institutions

○ Privacy and Data Ethics

■ Differential privacy and secure multi-party computation in sensitive data sharing

■ Bias and fairness in automated enforcement and account moderation

○ Societal Impact and Policy

■ Aligning AI defense with global regulations (e.g., EU AI Act)

■ Measuring the systemic risk of AI-driven abuse in democratic processes

● Open Science and Benchmarking

○ Developing verifiable, non-proprietary benchmarks for fraud and abuse

○Techniques for 'Privacy-Preserving Competition': Benchmarking on sensitive/secret data

○ Methodologies for validating SoTA claims without compromising corporate 'secret sauce'

Submission Guidelines

• Format: All submissions must be PDFs in the Standard ACM Conference Proceedings Template (sigconf format).

• Page Limit: 4–8 content pages (including figures/tables), excluding references.

• Anonymity: Reviews are double-blind. Submissions must not list author names or affiliations.

• Publication: Accepted papers will be posted on the workshop website. Note that KDD workshop

Papers are generally non-archival to allow for future journal submission.

Submission Portal (CMT3): https://cmt3.research.microsoft.com/AIFA2026/

Key Deadlines (all times are AoE midnight):

• Submission Deadline: April 30, 2026 May 7, 2026 DEADLINE EXTENDED

• Author Notification: June 4, 2026 June 11, 2026 TIMELINE REVISED

• Final Materials Due: June 22, 2026

• Workshop Date: August 9, 2026 (Tentative)