2024-08-07 –, Florentine F
Are you tired of the same old cybersecurity conference talks? Fed up with the routine
discussions about securing AI? Then get ready for something refreshingly different. Join me for
a quick adventure filled with offbeat anecdotes and outrageous scenarios – imagine
cybercriminals attempting to teach self-driving cars the cha-cha slide and chatbots gossiping
about their creators' music taste. Amidst the puns and dad jokes, this talk will unveil everything
you need to know about security for AI, including unconventional strategies to secure AI against
the unexpected. I'll do my best to keep you entertained every step of the way during this 101 talk.
This talk is a concise primer on AI fundamentals and common security challenges. It's meant to demystify AI concepts and discuss prevalent security threats, equipping attendees with essential knowledge. Attendees will be able to walk away with a clear understanding of how to navigate the intersection of AI and security effectively... and I'll try to make sure to bring some comedy to it.
Introduction
a. Pervasive use of AI data
b. Challenges in securing AI data
c. Security breaches looming data
d. Sources of AI breaches and current security measures data
e. Collaboration and concerns
f. Adversarial AI over time (timeline of 2002 - 2024)
Part 1: Risks Related to the Use of AI
a. Harmful content creation
b. Deep Fakes
c. Data Privacy and Leakage
d. Copyright violation
e. Accuracy and bias issues
f. Other ethical and societal issues
Part 2: Risks Faced by AI- based systems
a. Adversarial machine learning attacks
b. Attacks specific to generative AI
c. Supply chain attacks
d. Threat actors and attack vectors
Part 3: Advancements in Security for AI
a. Offensive security tooling for AI
b. Defensive frameworks for AI
c. Red Teaming and risk assessment
d. Policies and regulations
Part 4: Predictions and Recommendations
Predictions
a. Data scientists will partner with security practitioners to secure their models
b. Supply chain attacks using ML artifacts will become much more common
c. There will be a significant increase in adversarial attacks against AI
d. Threat actors will automate hacking efforts with LLMs
e. Deepfakes will be increasingly used in scam and disinformation
f. AI attack surfaces will expand while more organizations use advanced tools to
combat threats
Conclusion
a. Recommendations
b. Take home checklist to showcase where your organization stands when it comes to
securing their AI
Chloé Messdaghi serves as the Head of Threat Intelligence at HiddenLayer, where she spearheads efforts to fortify security for AI measures and fosters collaborative initiatives to enhance industry-wide security practices for AI. A highly sought-after public speaker and trusted authority for national and sector-specific journalists, Chloé's expertise has been prominently featured across various media platforms. Her impactful contributions to cybersecurity have earned her recognition as a Power Player by esteemed publications such as Business Insider and SC Media.Beyond her professional endeavors, Chloé remains passionately committed to philanthropy aimed at advancing industry progress and fostering societal and environmental well-being.