Breaking AI Security & Using LLM to Fuzzing Meetup

Expand your horizons by joining our meetup:

Leverage LLMs to take fuzzing to the next level (Arad Cohen)
Arad Cohen will introduce a novel methodology that harnesses large language models (LLMs) to extend the reach of fuzzing. By generating smarter test cases, the method improves code coverage in TLS libraries such as OpenSSL and raises the odds of uncovering hidden security flaws in critical software.

Building and Breaking AI Security (Amit Levi and Rom Himelstein)
Amit Levi and Rom Himelstein will outline their latest research on risks inside language models. They will show how weaknesses can crash systems or leak sensitive data, walk through real attack methods on deep learning models, and discuss the impact when attackers use the same tactics to control drones, robots, and autonomous vehicles

Additional information:
⚪ Date: June 15th, 17:30
⚪ Location: Tech&Talk MATAM, Haifa (Tech&Talk Matam is located right next to Gate 2 of Matam Park, easy to reach by train or public transportation)
⚪ Parking: Parking is available in the dirt lot just outside the park, near Intel's building
⚪ Language: The event will be held in Hebrew
⚪ Background: Basic understanding of TLS and how LLMs work

Agenda:
17:30 - Gathering, Networking and Pizzas 🍕🍕
18:00 - Leverage LLMs to take fuzzing to the next level
18:40 - Break
19:00 - ⁠Building and Breaking AI Security
19:40 - Q&A Panel
20:00 - End