Effective Altruism Osnabrück is happy to host Jan Kirchner, for a talk on the 26th of January at 7 pm. After finishing his PhD in computational neuroscience at the MPI, Jan is now a research engineer at OpenAI, one of the worlds leading AI labs.
His talk will be an introduction to technical aspects of the AI alignment problem - the problem of designing AI systems such that they are trying to do what their operators want them to do. He'll explain why failing to solve the alignment problem might pose risks to humanity and outline current attempts to solve the problem.
No background in AI is needed to follow this talk.