Cátedra ENIA UAM-VERIDAS en IA Responsable

(Code: TSI-100927-2023-2)

Description

The rapid advance of AI technologies, e.g., Generative AI and Foundation Models, opens the doors to the exploration and proposal of novel AI solutions in key scenarios of society. These novel solutions can enhance not only the performance but also other critical aspects of the current AI technology such as the explainability, bias, and privacy, following the recent European AI Act guidelines. Cátedra ENIA UAM-VERIDAS en IA Responsable focuses on the following main objectives: i) proposal of novel methods to detect and mitigate the extraction/use of sensitive information in AI models, ii) proposal of novel methods to reduce the bias in AI models, iii) proposal of novel methods to improve the explainability of the AI models, and iv) proposal of novel methods to increase the robustness and security performance of AI systems against attacks such as DeepFakes, poisoning and backdoor attacks, etc.

 

News