Carnegie Mellon University: School of Computer Science: Computer Science
Location: Pittsburgh, PA
Description:
Carnegie Mellon University's new CoRe Safe AI has an opening for a postdoctoral position on Conceptual Research on Safe AI. Research for this position focuses on developing conceptually new approaches to AI safety, informed by the latest advances but with an eye towards the future. The aim is to identify important aspects that are not yet visible today but that we should expect in the (possibly near) future.
The ideal candidate has demonstrated interest and ability to contribute to such topics. A variety of backgrounds and focus areas are potentially appropriate; strong backgrounds in fields including for example AI safety and alignment, decision theory and game theory, formal epistemology, and learning theory are all relevant. (It is not required to have a background in every one of these fields, and both experimental and theoretical methods are in scope.) Previous work on, or closely related to, the specific topics above is also highly desirable.
The successful candidate is able to quickly get into new research areas and will be responsible for actively engaging in novel research questions, as well as publishing and communicating research results. The candidate will also take a broader role in CoRe Safe AI, including collaborating with and possibly advising other members, and generally participating in and contributing to activities in the lab.
We expect the position to last 1 or 2 years (in the latter case, conditional on satisfactory review); it may be possible to extend the position beyond that. The starting date is flexible; please indicate constraints or preferences.
Qualifications:
PhD in Computer Science or related field such as Philosophy, Economics (Theory), or Mathematics.
