Postdoctoral Fellow – Costello Lab – Dietrich College
Core
Researching the risks of AI-driven persuasion within the information ecosystem and developing scalable interventions to mitigate these risks.
Role type
Postdoctoral Fellow (AI Safety & Behavioral Science)
Builds
Scalable interventions and tools for evaluating AI persuasive capabilities
Domain
AI safety, behavioral science, large language models (LLMs)
Deliverable
production ML models | research
Required skills
Large language model expertise (fine-tuning, prompting, evaluation), experimental design for belief change and persuasion, building user-facing tools, running interventions at scale
Preferred skills
Bridging technical/computational backgrounds with computational social science
Technologies
Large language models (LLMs)
Responsibilities
Designing and running experiments to examine how AI systems shape human beliefs; developing rigorous methods to evaluate and mitigate persuasive risks; creating interventions capable of operating at scale
Seniority
Postdoctoral Researcher