Exploring the Security and Safety of AI (WiCS seminar week 1)

Описание: Behind the Code: Exploring the Security and Safety of AI in Our Daily Lives

How often do you use tools like ChatGPT? How much do you trust the answers that it gives to you? In our daily lives, artificial intelligence is becoming increasingly important, but with its rise comes important questions about its safety and security. In this seminar we will break down Large Language Models (LLMs) to understand how they generate text and make decisions behind the scenes. With this understanding we will explore what can go wrong. What are the potential dangers that come with AI? You will see real-life examples of how vulnerabilities can affect AI-driven applications and even impact society. Then, we will discuss ways to build more secure and resilient LLMs, highlighting the current challenges and the big questions that still need answers in this field. Finally, we'll take a step back and think about the future: What ethical considerations should you keep in mind as AI continues to evolve? How can we make sure that AI technologies grow safely alongside us? Join us for an engaging and eye-opening session that not only covers the technical aspects of AI security but also encourages you to think deeply about the role of AI in your life and the world ahead.

This event was part of the Women in Computer Seminar series and took place as a live Zoom webinar on Tuesday 18 February 2025.