This interactive feature by The Straits Times highlights the privacy risks of AI chatbots. It also features research from ETH Zurich: Professor Martin Vechev and doctoral student Robin Staab from the Secure, Reliable, and Intelligent Systems lab show how models can infer surprisingly accurate personal traits from just a few text prompts. Read more.
A newly convened UN panel has launched a global study on the impact of artificial intelligence on society, the economy and human rights. Menna El-Assady, a Professor of Computer Science at ETH Zurich, is a member of the panel and contributes her expertise on responsible and transparent AI to its work. Read more.
The article reports on the accidental leak of source code from Anthropic’s coding assistant Claude Code, raising broader questions about security practices in the AI industry. It also cites ETH Zurich researcher Professor Florian Tramèr, who warns that fast-moving AI startups must strengthen their security culture as sensitive data is increasingly entrusted to such systems. Read more (in German).