Postdoc, Mechanism Design for Robust AI Alignment

Carmine Ventre’s Post

• Report this post

𝗗𝗼 𝘆𝗼𝘂 𝘄𝗮𝗻𝘁 𝘁𝗼 𝗵𝗲𝗹𝗽 𝗺𝗲 𝗯𝗿𝗶𝗱𝗴𝗲 𝗺𝗲𝗰𝗵𝗮𝗻𝗶𝘀𝗺 𝗱𝗲𝘀𝗶𝗴𝗻 𝗮𝗻𝗱 𝗔𝗜 𝗮𝗹𝗶𝗴𝗻𝗺𝗲𝗻𝘁? Our project, "The Implementability-Capability Frontier: Mechanism Design for Robust AI Alignment," has been awarded funding through the UK AI Security Institute and UK Research and Innovation ! This work aims to lay a theoretical foundation for aligning advanced AI systems using mechanism design—asking: What can we guarantee about AI behaviour when agents are powerful and have evolving capabilities? We are building a team at King's College London to tackle this challenge and are now seeking a postdoc to join us. 🔍 𝗪𝗲’𝗿𝗲 𝗹𝗼𝗼𝗸𝗶𝗻𝗴 𝗳𝗼𝗿 𝘀𝗼𝗺𝗲𝗼𝗻𝗲 𝘄𝗶𝘁𝗵: - A PhD in computer science, economics, operations research, or a related field. - Strong background in mechanism design, algorithmic game theory, or social choice theory. - A strong interest in applying theoretical tools to AI alignment challenges. If you enjoy thinking about incentive compatibility, solution concepts beyond dominant strategies, imperfect rationality, and how these ideas apply to the alignment of frontier AI systems—I’d love to hear from you. 𝗜𝗻𝘁𝗲𝗿𝗲𝘀𝘁𝗲𝗱? Please send me a direct message with your CV and a brief note on how your background fits. A formal call will follow, but I’m keen to start the conversation early with the right people. This is a chance to do foundational work in one of the most important research areas of our time—with the support of a world-class institute and the academic freedom to explore bold ideas.