Navigating the Risks of AI Language Models

by Raju Kansadwamy

Raju Kansadwamy

As we navigate through the ChatGPT era, users' tendencies are increasingly shifting towards utilising conversational interfaces driven by large language models (LLM). But, what does this transition mean for security? This session aims to provide an experiential understanding of the security concerns tied to conversational AI models. We'll delve into concepts such as prompt injection and hijacking, the phenomenon of 'jailbreaking', and the ethical and anti-social considerations that emerge from these new technologies.

Raju Kansadwamy

Principal Consultant, Thoughtworks

Innovative tech enthusiast and strategist with experience in XR and industrial robotics. Skilled in building and upscaling autonomous teams. Successfully helped a top IT hardware brand design and develop software for their mixed reality glass product, resulting in 6 filed patents. Built a tech start-up from the ground up and successfully handled assignments in setting up new product development teams. Skilled in real-time high-transaction rate enterprise systems and managing development teams.