The literature review is no longer a meaningful student assignment. Neither, arguably, is the annotated bibliography. Generative AI tools complete both with speed and competence that make them poor vehicles for assessing student learning. But if these familiar assignments have run their course, what replaces them, and what does that mean for how we teach in the social sciences?
The answer lies in recovering what research design actually is: not a set of technical procedures, but a sequence of deliberate analytical decisions made under conditions of uncertainty, constraint, and ethical responsibility. When we teach it that way, AI becomes a resource to think with rather than a shortcut that renders assessment meaningless.
The proposed session is organized around three interconnected themes. First, we examine how to intentionally integrate AI tools at specific stages of the research design process. Problem formulation, literature mapping, comparison of methodological options, and data collection instrument design each present distinct opportunities for AI-assisted learning, as well as distinct risks.
Second, we consider how to selectively develop research design components for different student populations and purposes. The analytical demands of undergraduate versus graduate work differ meaningfully, as do scholarship-oriented versus practice-oriented research. AI tools can be calibrated accordingly, or misused in ways that flatten those distinctions.
Third, the session makes the case for research question development as the most productive site of AI-resistant, process-oriented learning and assessment across the social sciences. Formulating a good research question requires iteration, judgment, and disciplinary knowledge that AI cannot substitute. Centering assessment here gives students and instructors a foothold that generative tools cannot easily erode.
This session is designed for social science faculty at any level, including those teaching methods, theory, area studies, or applied policy, who are rethinking how to design assignments and assessments that elevate critical thinking and human judgment when AI is already in the room. Attendees will leave with a practical framework for intentionally integrating AI at specific stages of the research design process, strategies for calibrating that integration to different student populations and course purposes, and a concrete approach to centering research question development as an AI-resistant site of deep learning.
