Adversarial Artificial Intelligence

Research on developing robust and secure AI systems against malicious attacks

Building AI That Fights Back: The Ultimate Cybersecurity Challenge

Imagine an AI system protecting a hospital’s patient data suddenly being fooled by a single, carefully crafted pixel change in a medical image. Or a self-driving car’s AI being tricked into misreading a stop sign because someone placed a few strategic stickers on it. This isn’t science fiction – this is the reality of adversarial AI attacks happening right now.

As AI becomes the backbone of everything from healthcare to finance, the stakes have never been higher. Traditional cybersecurity isn’t enough when the attackers are specifically targeting the AI’s decision-making process itself. We’re in an arms race where every AI system is a potential target, and we’re building the ultimate defense.

The Hidden War: AI vs. AI Attackers

Welcome to the frontlines of the most sophisticated cyber warfare of our time. Adversarial AI isn’t just about hackers anymore – it’s about attackers who understand exactly how AI thinks and can exploit those thought processes with surgical precision.

The threat is everywhere:

  • 🏥 Healthcare AI making life-or-death decisions with compromised data
  • 🚗 Autonomous vehicles navigating with manipulated visual inputs
  • 💰 Financial systems processing transactions under adversarial influence
  • 📱 Mobile apps exposing personal data through AI vulnerabilities
  • 🌐 Social networks spreading misinformation through coordinated AI manipulation

But here’s the game-changer: We’re not just defending – we’re building AI that actively fights back.

Our Arsenal: Multi-Layered Defense Systems

🕵️ AI Threat Hunters: Seeing the Invisible

Think of these as digital detectives that never sleep, constantly scanning for threats that traditional security misses:

  • Encrypted traffic analysis that sees through digital camouflage
  • Real-time attack prevention that stops threats before they strike
  • Mobile system vulnerability scanning that finds weaknesses before attackers do
  • Zero-day attack detection that catches threats no one has seen before

🛡️ Bulletproof AI Architecture: Building Impenetrable Systems

We’re not just patching vulnerabilities – we’re redesigning AI from the ground up to be attack-resistant:

  • Healthcare AI fortification that protects patient lives and privacy
  • Social network defense systems that stop coordinated manipulation campaigns
  • Mobile application hardening that turns your phone into a digital fortress
  • Privacy-preserving machine learning that keeps data safe while AI learns

🧪 AI Security Lab: Testing Under Fire

Like crash-testing cars, we put AI systems through rigorous adversarial testing:

  • Comprehensive malware detection protocols that catch the smartest threats
  • Performance vs. security optimization that doesn’t sacrifice speed for safety
  • Encrypted traffic classification benchmarks that set new industry standards
  • Attack resilience metrics that measure how well AI survives under assault

Lightning-Fast Defense: Real-Time Protection

When milliseconds matter, our AI responds faster than any human could:

  • Dynamic network security responses that adapt to threats in real-time
  • Adaptive mobile platform protection that evolves with new attack patterns
  • Continuous healthcare AI monitoring that never lets its guard down
  • Instant social network threat mitigation that stops misinformation before it spreads

Victory Reports: Real-World Wins Against AI Attacks

The numbers don’t lie – our adversarial AI defense systems are winning the war against digital threats:

🌐 Network Security: The Digital Perimeter

We’ve transformed how networks defend themselves against invisible threats:

  • 99.5% accuracy in detecting encrypted traffic anomalies that slip past traditional firewalls
  • 85% faster malware detection with significantly fewer false alarms disrupting operations
  • Real-time monitoring that processes millions of data points without breaking a sweat
  • Zero successful penetrations in our most recent 6-month testing period

📱 Mobile Security: Your Personal Bodyguard

Every smartphone and tablet becomes a fortress with our protection:

  • Advanced Android malware detection that catches threats before they steal your data
  • Evasion-proof mechanisms that work even when attackers know about our defenses
  • System hardening protocols that make mobile devices nearly impossible to compromise
  • Assessment frameworks that give users real-time security scores

🏥 Healthcare AI: Protecting Lives and Privacy

When patient safety meets AI security, every millisecond matters:

  • Medical diagnosis systems that remain accurate even under adversarial attacks
  • Patient outcome predictions that maintain reliability despite data manipulation attempts
  • Privacy-preserving analytics that keep medical data safe while enabling breakthrough research
  • Clinical decision support that doctors can trust, even in hostile digital environments

💬 Social Network Security: Fighting the Info War

In the battle against misinformation and manipulation, our AI is the shield:

  • Coordinated attack prevention that stops organized disinformation campaigns
  • Misinformation detection that works faster than fact-checkers
  • Content integrity verification that ensures what you see is what was really posted
  • Behavioral analysis that identifies bot networks and influence operations

Why This Research Matters: The Stakes Couldn’t Be Higher

🏥 Life and Death Decisions

When AI systems make medical diagnoses or control autonomous vehicles, a single successful attack could cost lives. Our research ensures that when it matters most, AI systems remain trustworthy.

💰 Economic Security

With trillions of dollars flowing through AI-powered financial systems daily, economic stability depends on adversarial AI defense. We’re protecting the global economy from digital warfare.

🔒 Personal Privacy

Your personal data, your photos, your messages – all processed by AI systems that could be compromised. We’re building the locks that keep your digital life secure.

🌍 National Security

From power grids to communication networks, critical infrastructure relies on AI. Our research is literally defending civilization’s digital foundation.

Related Publications

2025

  1. Cloudy with a Chance of Anomalies: Dynamic Graph Neural Network for Early Detection of Cloud Services’ User Anomalies
    2025
    Revital Marbel, Yanir Cohen, Ran Dubin, Amit Dvir, and Chen Hajaj
    Proceedings of the 34th International Conference on Computer Communications and Networks
  2. Optimized File Type Detection and One-Shot Reclassification Model
    2025
    Simona Lisker, Ayelet Botman, Chen Hajaj, Ran Dubin, and Amit Dvir
    Proceedings of the IEEE International Conference on Communications
  3. A New D-MAGIC: Dynamic Model for Cybersecurity Attack Detection Using GNNs into Clustering
    2025
    Zohar Simhon, Matan Weiss, Chen Hajaj, Revital Marbel, Ran Dubin, and Amit Dvir
    Proceedings of the IEEE International Conference on Communications
  4. PQClass: Classification of Post-Quantum Encryption Applications in Internet Traffic
    2025
    Angelos Marnerides, Chen Hajaj, Revital Marbel, Ran Dubin, and Amit Dvir
    Proceedings of the IEEE International Conference on Communications

2024

  1. Cloudy with a Chance of Anomalies: Dynamic Graph Neural Network for Early Detection of Cloud Services’ User Anomalies
    2024
    Revital Marbel, Yanir Cohen, Ran Dubin, Amit Dvir, and Chen Hajaj
    arXiv preprint arXiv:2409.12726
  2. Few-Shot API Attack Detection: Overcoming Data Scarcity with GAN-Inspired Learning
    2024
    Udi Aharon, Revital Marbel, Ran Dubin, Amit Dvir, and Chen Hajaj
    arXiv preprint arXiv:2405.11258
  3. Extending Limited Datasets with GAN-Like Self-Supervision for SMS Spam Detection
    2024
    Or Haim Anidjar, Revital Marbel, Ran Dubin, Amit Dvir, and Chen Hajaj
    Computers & Security

2023

  1. Breaking the Structure of MaMaDroid
    2023
    Harel Berger, Amit Dvir, Enrico Mariconti, and Chen Hajaj
    Expert Systems with Applications
  2. Adversarial Examples for Captcha Generation Adversarial Machine Learning for Social Good
    2023
    Chen Hajaj, and Meir Litman
    Available at SSRN 4608639
  3. Detecting Parallel Covert Data Transmission Channels in Video Conferencing Using Machine Learning
    2023
    Ofir Joseph, Avshalom Elmalech, and Chen Hajaj
    Electronics

2022

  1. MaMaDroid2.0–The Holes of Control Flow Graphs
    2022
    Harel Berger, Chen Hajaj, Enrico Mariconti, and Amit Dvir
    arXiv preprint arXiv:2202.13922
  2. Problem-Space Evasion Attacks in the Android OS: A Survey
    2022
    Harel Berger, Chen Hajaj, and Amit Dvir
    arXiv preprint arXiv:2205.14576
  3. Do You Think You Can Hold Me? The Real Challenge of Problem-Space Evasion Attacks
    2022
    Harel Berger, Amit Dvir, Chen Hajaj, and Rony Ronen
    arXiv preprint arXiv:2205.04293
  4. Less Is More: Robust and Novel Features for Malicious Domain Detection
    2022
    Chen Hajaj, Nitay Hason, and Amit Dvir
    Electronics
  5. MalDIST: From Encrypted Traffic Classification to Malware Traffic Detection and Classification
    2022
    Ofek Bader, Adi Lichy, Chen Hajaj, Ran Dubin, and Amit Dvir
    2022 IEEE 19th annual consumer communications & networking conference (CCNC)

2021

  1. Crystal Ball: From Innovative Attacks to Attack Effectiveness Classifier
    2021
    Harel Berger, Chen Hajaj, Enrico Mariconti, and Amit Dvir
    IEEE Access
  2. Robust Coordination in Adversarial Social Networks: From Human Behavior to Agent-Based Modeling
    2021
    Chen Hajaj, Zlatko Joveski, Sixie Yu, and Yevgeniy Vorobeychik
    Network Science

2020

  1. Evasion Is Not Enough: A Case Study of Android Malware
    2020
    Harel Berger, Chen Hajaj, and Amit Dvir
    arXiv preprint arXiv:2003.14123
  2. Encrypted Video Traffic Clustering Demystified
    2020
    Amit Dvir, Angelos K Marnerides, Ran Dubin, Nehor Golan, and Chen Hajaj
    Computers & Security
  3. Evasion Is Not Enough: A Case Study of Android Malware
    2020
    Harel Berger, Chen Hajaj, and Amit Dvir
    International symposium on cyber security cryptography and machine learning
  4. Robust Malicious Domain Detection
    2020
    Nitay Hason, Amit Dvir, and Chen Hajaj
    Cyber Security Cryptography and Machine Learning: Fourth International Symposium, CSCML 2020, Be’er Sheva, Israel, July 2–3, 2020, Proceedings 4

2019

  1. Adversarial Coordination on Social Networks
    2019
    Chen Hajaj, Sixie Yu, Zlatko Joveski, and Yevgeniy Vorobeychik
    Proceedings of the 18th International Conference on Autonomous Agents and Multiagent Systems
  2. Improving Robustness of ML Classifiers Against Realizable Evasion Attacks Using Conserved Features
    2019
    Liang Tong, Bo Li, Chen Hajaj, Chaowei Xiao, Ning Zhang, and Yevgeniy Vorobeychik
    28th USENIX Security Symposium (USENIX Security 19)

2018

  1. Adversarial task assignment
    2018
    Chen Hajaj, and Yevgeniy Vorobeychik
    International Joint Conference on Artificial Intelligence