Stanislav Fořt: Robustnost a adversariální útoky

It is necessary to register for the event here: [https://www.google.com/url?q=https://mesic.ai/akce/jake-zaludnosti-muze-prinest-ai-robustnost-a-adversarialni-utoky /](https://www.google.com/url?q=https%3A%2F%2Fmesic.ai%2Fakce%2Fjake-zaludnosti-muze-bring-ai-robustnost-a-adversarialni-attacks%2F&fbclid=IwZXh0bgNhZW0CMTAAAR2699N_Es_16HIcNGZFDSAKBzlJfl7c0m9PqbcOG9h5K0ab8zKS paFV3A4_aem_md -tG-jgcsdrhm1QteXNhw)

This seminar with Stanislav Forte will focus on the robustness of AI systems and so-called adversarial attacks that can confuse AI systems and threaten their reliability. You will learn the risks of these attacks and how to defend against them.

Stanislav Fořt is a recognized expert in the field of artificial intelligence, focused on large language models and AI security issues. During his career, he led research and development of open-source language models for Stability AI, participated in the development of the Claude system at Anthropic, and also worked on the Google Brain and DeepMind teams. He received his doctorate in artificial intelligence from Stanford University in the US, while he studied physics at the University of Cambridge. He has published more than 30 professional articles.

You'll gain deeper insight into how AI models can be vulnerable to adversarial attacks, where seemingly innocuous input changes can completely change their outputs. You will learn how to ensure the robustness of your AI systems and how to minimize the risks associated with their breach.
Program:

18:30 – 19:00: Open door and registration
19:00 – 20:30: Presentation by Stanislav Fořt
20:30 – 21:00: Networking

We are co-organizing the event with the Czech Association of Artificial Intelligence as part of AI Month. Show less