AI Pioneer Yoshua Bengio to Visit TU Berlin
Public lecture by AI pioneer Yoshua Bengio on AI safety, the risks of autonomous agents, and new approaches to developing safe-by-design artificial intelligence.
The Technische Universität Berlin and the Berlin Institute for the Foundations of Learning and Data (BIFOLD) welcome Yoshua Bengio, Turing Award recipient and one of the pioneers of modern AI, for a public lecture on the safety challenges posed by increasingly capable artificial intelligence.
Bengio will discuss recent AI safety incidents involving autonomous agents, their potential real-world consequences, and the technical and societal safeguards needed to address emerging risks. A central topic is the AI alignment problem: how to ensure that AI systems act in accordance with human intentions, values, and safety requirements.
The lecture will also introduce research at LawZero, the nonprofit organization co-led by Bengio, which aims to develop demonstrably safe-by-design AI. Its Scientist AI approach seeks to create systems without hidden goals or preferences, emphasizing interpretable, auditable, and verifiable predictions.
- Date: November 6, 2026
- Time: 5:00–7:00 PM
- Location: TU Berlin, Mathematics Building, Room MA001, Straße des 17. Juni 136, 10623 Berlin
- Organizer: Berlin Institute for the Foundations of Learning and Data (BIFOLD)
- Speaker: Prof. Yoshua Bengio, Turing Award recipient, Professor at Université de Montréal, Co-President and Scientific Director of LawZero
- Admission: Free, mandatory registration; admission only with registration confirmation
- Language: English
- Audience: Researchers, students, AI developers, and anyone interested in AI safety
- Focus: AI safety, autonomous agents, AI alignment, safe-by-design AI, Scientist AI, technical safeguards, and AI governance
- Program: Public lecture on AI safety challenges and LawZero's research approaches