Generative Artificial Intelligence (GenAI) has rapidly become an influential research-support technology, enabling researchers to generate text, summarize information, refine language, assist with ideation, and support other stages of scholarly work. Its increasing integration into research processes, however, introduces ethical concerns involving authorship, transparency, accountability, originality, privacy, bias, misinformation, and responsible use. This article develops a conceptual framework for evaluating researchers’ ethical perceptions and practices toward GenAI. The study adopts a research-and-review approach based exclusively on the supplied literature concerning artificial intelligence ethics, ChatGPT, academic research, publication ethics, higher education, and responsible scholarly use. The proposed framework integrates five principal dimensions: ethical awareness, perceived risks and benefits, responsible-use practices, transparency and accountability, and institutional governance. The analysis indicates that ethical awareness should not be understood merely as knowledge of ethical principles; rather, it should be evaluated through the relationship between researchers’ perceptions and their actual practices when using GenAI. The framework further distinguishes between awareness and responsible behavior, emphasizing that researchers may recognize ethical risks while still lacking consistent mechanisms for disclosure, verification, or accountability. The article concludes that systematic ethical evaluation can strengthen responsible GenAI adoption in research environments by connecting individual researcher judgment with institutional policies and professional standards.