As artificial intelligence (AI) becomes more available and sophisticated, malicious actors are increasingly turning to AI-powered tools. From automated phishing campaigns and deepfakes to hostile malware, malicious actors are leveraging the power of AI to defeat traditional defenses.
AI also enables a more proactive rather than reactive security posture. The solution continuously monitors networks and endpoints, thwarting threats and alerting teams to suspicious behavior that may represent the early stages of the intrusion lifecycle.
This is the crux of what I will be speaking about at RSA 2024 in the session “AI-powered threat actors vs. AI-enhanced cyber tools – who will win?” As the leader of BlackBerry's product engineering and data science teams, I'm excited to share the progress we've made to enhance our Cylance® AI-powered solutions.
Our researchers have discovered new tactics from several advanced persistent threat groups targeting critical infrastructure. Through detailed analysis of these attacks, our data scientists enhanced the Cylance AI model to more accurately identify malicious behavior and tools.
I would like to share a quick preview of my session. We hope you'll join her for a lively Q&A at RSA in May.
AI in Cybersecurity: Future Challenges and Opportunities
By developing AI-enhanced detection and response tools, defenders can gain insights to identify emerging threats. Machine learning models can analyze vast amounts of data at machine speed and detect subtle anomalies and patterns that can signal the beginning of an attack. Machine learning models trained to look for malicious intent do a great job of identifying new and invisible suspicious behavior.
However, AI systems are not foolproof. We show that an adversary can manipulate the inferences of his ML (machine learning) model through small perturbations in the input data and evade detection. Defenders must take precautions to minimize these risks and protect their AI tools.
My RSA presentation will cover:
-
Examining data science and modeling tools that threat actors may use or are using to create targeted attacks leveraging ML techniques.
-
Approaches defenders can take to address the rise in AI/ML-based threat detection
-
Learn about adversarial attacks on the ML model itself and how to reduce the risk.
-
Explore predictive and behavioral modeling, a powerful tool for defenders, and how they are solving the challenges they face.
Cyber defenders have the tools to fight back in the AI arms race, but only if they implement strategies that minimize risk and protect their systems. A balanced and carefully managed approach that combines the strengths of threat research and AI may be defenders’ best hope for keeping pace with malicious attackers in the long run.
