Rising publication pressure and the routine use of generative AI tools are reshaping how software engineering research is produced, assessed, and taught. While these developments promise efficiency, they also raise concerns about skill degradation, responsibility, and trust in scholarly outputs. This vision paper employs Design Fiction as a methodological lens to examine how such concerns might materialise if current practices persist. Drawing on themes reported in a recent community survey, we construct a speculative artifact situated in a near future research setting. The fiction is used as an analytical device rather than a forecast, enabling reflection on how automated assistance might impede domain knowledge competence, verification, and mentoring practices. By presenting an intentionally unsettling scenario, the paper invites discussion on how the software engineering research community in the future will define proficiency, allocate responsibility, and support learning.