As generative AI (GenAI) is increasingly applied in persona development to represent real users, understanding the implications and limitations of this technology is essential for establishing robust practices. This scoping review analyzes how 81 articles (2022-2025) use GenAI techniques for the creation, evaluation, and application of personas. The articles exhibited good level of reproducibility, with 61% of articles sharing resources (personas, code, or datasets). Furthermore, conversational persona interfaces are increasingly provided alongside traditional profiles. However, nearly half (45%) of the articles lack evaluation, and the majority (86%) use only GPT models. In some articles, GenAI use creates a risk of circularity, in which the same GenAI model both generates and evaluates outputs. Our findings also suggest that GenAI seems to reduce the role of human developers in the persona-creation process. To mitigate the associated risks, we propose actionable guidelines for the responsible integration of GenAI into persona development.

Recent work on large language models (LLMs) has increasingly focused on post-training and alignment with datasets curated to enhance instruction following, world knowledge, and specialized skills. However, most post-training datasets used in leading open- and closed-source LLMs remain inaccessible to the public, with limited information about their construction process. This lack of transparency has motivated the recent development of open-source post-training corpora. While training on these open alternatives can yield performance comparable to that of leading models, systematic comparisons remain challenging due to the significant computational cost of conducting them rigorously at scale, and are therefore largely absent. As a result, it remains unclear how specific samples, task types, or curation strategies influence downstream performance when assessing data quality. In this work, we conduct the first comprehensive side-by-side analysis of two prominent open post-training datasets: Tulu-3-SFT-Mix and SmolTalk. Using the Magpie framework, we annotate each sample with detailed quality metrics, including turn structure (single-turn vs. multi-turn), task category, input quality, and response quality, and we derive statistics that reveal structural and qualitative similarities and differences between the two datasets. Based on these insights, we design a principled curation recipe that produces a new data mixture, TuluTalk, which contains 14% fewer samples than either source dataset while matching or exceeding their performance on key benchmarks. Our findings offer actionable insights for constructing more effective post-training datasets that improve model performance within practical resource limits. To support future research, we publicly release both the annotated source datasets and our curated TuluTalk mixture.