Most companies are run by software. Software undergirds everything from heath care to critical infrastructure. As software enables modern life and drives productivity, it also creates an expanding attack surface for malicious actors.

Like humans, AI systems are designed to learn from their experience. Unlike most humans, however, AIs aren’t constrained by norms, values, and assumptions. This means they can devise novel attacks that humans would never consider.

In the grand push for data, AI, and machine learning, we’ve entirely forgotten that we are still figuratively throwing ourselves into those machines. What comes out, and what comes of it, is only what we have enabled. That enablement might be based on a mistake, ignorance, willful ignorance, or malice.

As an industry, we’re so focused on gathering more data that we haven’t spent enough time making sure we’re using the right processes to gather the right data, protect it, and use it in the right way.

We’re not yet at a place where machines can make all the advanced inferences and detect the nuances that humans can. As models are built after the data – and the people working with the data should be protected.

For AI systems to function properly, a continuous feed of quality, trusted data is required. When bad, distorted, or incomplete data creeps into models it will make the false positive problem we currently face much worse. We can expect attackers to exploit the vulnerabilities associated with AI models, which can erode the trust.

In this talk we will cover:

• Attack surface of AI and Gen AI systems
• Most common Threats against AI systems. We will cover MITRE ATLAS
• Better manage the risks of AI systems and impact on a security program
• Learnings and fundamental precautions from existing security programs can be applied to protect LLM Systems
• AI Bill of Materials (AIBoM) and AI supply chain risks

Speaker: Srinivas Mukkamala PhD, Chief Product Officer, Ivanti

Dr. Srinivas Mukkamala is one of the world’s preeminent authorities on cybersecurity, artificial intelligence, and neural networks. He currently serves as Chief Product Officer for Ivanti, where he is responsible for Product Management and Engineering for all Ivanti products. Dr. Mukkamala and Ivanti are serving a large global customer base that includes 88 of the Fortune 100 companies and customers in 90+ countries as well as several Federal departments.

Dr. Mukkamala is known for his expertise at the intersection of cybersecurity and artificial intelligence, as an early adopter of these technologies. And put his expertise to work as the founder of RiskSense, the recognized leader in Risk-Based Vulnerability Management. Ivanti acquired RiskSense in 2021.

He has written and contributed to countless publications (with 6,000+ citations and counting) and collaborated with the U.S. Department of Defense to address high-stakes cybersecurity challenges.

Dr. Mukkamala holds a Ph.D. and a Master of Science in Computer Science from New Mexico Institute of Mining and Technology, one of the nation’s top technical institutions. He now serves as a board of regents of New Mexico Tech, and on El Paso Electric Company’s Board of Directors, among many other advisory roles.

CPE

1 CPE point