Snapshot: A Next-Generation Social Media Platform for Dynamic, Evolving Memories Using Artificial Intelligence and Collaborative Storytelling

This paper presented Snapshot, a conceptual framework for rethinking social media as a space for evolving memories rather than static posts or disappearing stories. Built on ideas from research in social media, human-computer interaction, and artificial intelligence, Snapshot responds to common issues like fragmented memories, shallow engagement, and the effects of algorithm-driven feeds on digital well-being. The system architecture, which includes the Frontend, Backend, and AI layers, shows how user experience, collaborative storytelling, and responsible AI can work together in one integrated model. The main contribution of this study is theoretical: it reimagines social media not only as a way to share content, but as a medium for collective memory-making and authentic interaction. Features such as Living Snapshots, time capsules, and mood-based feeds highlight how online experiences can become more meaningful and emotionally engaging. Although the framework remains conceptual, it points to a new direction for designing communication platforms that put people’s relationships and well-being at the center. Future work should move toward testing and validation. A prototype or minimum viable product (MVP) would make it possible to study Snapshot in practice and evaluate usability, feasibility, and impact. User studies across different age groups and cultural contexts could show how the platform shapes engagement, authenticity, and community. Collaboration between fields—such as computer science, psychology, sociology, and design—will be key to refining the model. Just as importantly, future research must explore the ethical use of AI, with attention to issues like transparency, privacy, and emotional influence. In summary, Snapshot represents an early step toward reimagining what social media can be. By blending technology with human-centered values, it offers a vision of platforms that do more than capture isolated moments—they help people and communities co-create evolving digital legacies.

1.      Haque, A. K. M. B., Islam, A. K. M. N., & Mikalef, P. (2025). To Explain Or Not To Explain: An Empirical Investigation of AI-Based Recommendations on Social Media Platforms. arXiv. This paper explores how users react when AI recommendations on platforms like Facebook are opaque vs. when they are explained clearly. arXiv   2.      Kalokyri, V., Borgida, A., & Marian, A. (2022). Supporting human memory by reconstructing personal episodic narratives from digital traces. Proceedings of the International AAAI Conference on Web and Social Media, 16(1), 453-464. This study shows how personal digital traces from multiple sources can be combined to help people recall their past in narrative form. AAAI Open Access   3.      Singh, S., & Tripathi, K. M. (2024). The impact of social media on social comparison and subjective well-being. International Journal of Indian Psychology, 12(1), 1152-1176. This paper finds that when people compare themselves to others a lot on social media, it tends to hurt their self-worth and mood. IJIP   4.      Beyari, H., & Hashem, T. (2025). The role of artificial intelligence in personalizing social media marketing strategies for enhanced customer experience. Behavioral Sciences, 15(5), 700. After surveying almost 900 people in the MENA region, the study confirms that personalization really matters—when done right. MDPI   5.      “Passive social media use and psychological well-being during the COVID-19 pandemic: The role of social comparison and emotion regulation.” (2022). Computers in Human Behavior, 127, Article 107050. This article looks at how people who passively scroll through social media—especially during stressful times like lockdowns—tend to compare themselves with others, which worsens their emotional state unless they use emotion regulation techniques. ScienceDirect   6.     Vachharajani, P. (2024). Enhancing social media personalization: Dynamic user profile embeddings and multimodal contextual analysis using transformer models. arXiv. This work shows that using dynamic embeddings (ones that update as users’ preferences change) leads to better recommendation accuracy and keeps engagement higher than static profiles. arXiv