February WiSKC

Arriving to Onspring after 6pm? The doors may be locked, just email info@wiskc.org and we'll let you in!

Topic: Can You Trust Your Tokens? Designing Secure Gen AI Systems

Generative AI runs on tokens—but trust isn’t automatic. Every token reflects decisions about data access, architecture, and risk made long before a response is generated. This session explores how to design secure Gen AI systems from the start: defining what must be protected, minimizing data exposure, building evaluation into development, and continuously testing against adversarial threats. 

Speaker: Liji Thomas

Speaker Bio: Liji Thomas is an accomplished technologist and leader with over 15 years of experience in building cutting-edge digital experiences. She is a Microsoft MVP in AI, which is a testament to her expertise and contributions to the community in the field of artificial intelligence. As Gen AI Manager at HRBlock, Liji Thomas oversees the development of cutting-edge solutions that leverage data and artificial intelligence to drive business value. Her expertise spans a wide range of technologies, including AI, machine learning, natural language processing, and conversational AI. In addition to her work at Valorem Reply, Liji is an active participant in the broader data and AI community. She is a frequent speaker at industry events and conferences.