Designing Social Media Futures

Why young people should design social media

Social media platforms are designed artifacts, and their designs carry consequences. Feeds, notifications, defaults, and interaction patterns shape how people — especially young people — connect, compare, present themselves, and build community. Yet in policy, education, and public discourse, social media is increasingly framed less as infrastructure for connection than as a threat from which young people must be protected. That framing is understandable, but it has a cost. Kim's research shows that an emphasis on harm can generate dysfunctional levels of privacy concern in young people: anxiety that degrades quality of life without meaningfully improving safety or leading to constructive action. Because peer relationships are developmentally central, exclusion from social media can also mean exclusion from peer life. Young people need opportunities to build resilience and shape healthier norms, not only rules that remove access.

This course grows out of a research program on designing social media that supports meaningful connection for young people. Three findings from that work set its agenda. First, design demonstrably shapes how young people present themselves. A study of BeReal with 29 teenage users found that its randomly timed notifications, retake counters, reciprocal posting requirements, and deliberately minimal interface effectively cultivated a norm of casual, uncurated sharing — proof that a platform's design, not merely its users, determines what authenticity looks like on it. But the same study documented the limits of that approach: teens described unfiltered photos at unplanned moments as only one facet of authentic self-presentation, and the platform's explicit quantification of authenticity sometimes produced toxic competition over who was being real enough.

Second, privacy and connection are not opposed, and designing them as opposites fails young people. A three-part study with 19 adolescents found that teens experience mundane, everyday sharing as meaningful self-disclosure, yet frequently withhold it. Two barriers recurred: communication fog, meaning ambiguous norms about loyalty, relevance, and what a reaction signifies, and low-grace culture, in which presentation expectations and the social risk of a misstep compound into cycles of distrust. The resulting framework, trust-enabled privacy, traces how platform design establishes norms, how those norms shape whether an audience reads as trustworthy, and how that judgment determines whether a young person shares, withdraws, or self-censors. Privacy, on this account, is a relational and developmental process rather than an individual burden or a binary switch.

Third, and most important for this course, the problem is often framed before young people are consulted. Kim's dissertation identifies three forms of what it calls problem-space misattunement. Conceptual misattunement occurs when the very phrase “social media” anchors participants to the interaction templates of the platforms they already know. In co-design workshops with 23 young people aged 15–24, participants initially asked for better social media in negative terms: less doomscrolling, fewer hours on reels, less performative self-presentation. Reframed as Hogwarts students imagining how to reach remote friends by magic, they described something else entirely — coexisting ambiently in a shared room while listening to music, houses that reflected their ideal selves, neighborhoods where closer friends lived closer by. Their visions resembled Minecraft or Discord more than Instagram. Definitional misattunement occurs when researchers and designers decide what “better” means on young people's behalf: youth are invited to generate solutions but rarely to define what the design is for. Evaluative misattunement occurs when people are asked to judge static or hypothetical designs whose real effects would emerge only over time and at scale.

Taken together, these findings make an argument this course acts on. Young people hold rich relational experience with social media, but the available language for describing it — platform interface labels, harm-centered public discourse, predefined research constructs — is borrowed and often ill-fitting. Involving young people in design is therefore necessary, and involvement alone is not sufficient: without care in how the problem is framed, participants reproduce the adult narratives they have been taught. Teaching undergraduates to read platform design critically, then asking them to build alternatives and defend the values embedded in them, is one way to widen the frame before it hardens.

Publications this course draws on

• Kim, J., Wolfe, R., Chordia, I., Davis, K., & Hiniker, A. (2024). “Sharing, Not Showing Off”: How BeReal Approaches Authentic Self-Presentation on Social Media Through Its Design. Proc. ACM Hum.-Comput. Interact. 8 (CSCW2).
• Kim, J., Wolfe, R., Subramanian, R. B., Lee, M.-H., Colnago, J., & Hiniker, A. (2025). Trust-Enabled Privacy: Social Media Designs to Support Adolescent User Boundary Regulation. SOUPS 2025.
• Kim, J., Cho, S., Wolfe, R., Nair, J. H., & Hiniker, A. (2025). Privacy as Social Norm: Systematically Reducing Dysfunctional Privacy Concerns on Social Media. Proc. ACM Hum.-Comput. Interact. 9 (CSCW).
• Kim, J., Cho, H., Liu, F., & Hiniker, A. Social Media Should Feel Like Minecraft, Not Instagram: Youth Visions for Meaningful Social Connections through Fictional Inquiry. arXiv:2502.06696.
• Kim, J., Klein-Balajee, T., Kelly, R. M., & Hiniker, A. Discord's Design Encourages “Third Place” Social Media Experiences. arXiv:2501.09951.
• Kim, J., Popowski, L., Fang, A., Pyle, C., Freeman, G., Kelly, R. M., Lee, A. Y., Liu, F., Smith, A. D. R., To, A., & Zhang, A. X. (2024). Envisioning New Futures of Positive Social Technology: Beyond Paradigms of Fixing, Protecting, and Preventing. CSCW Companion 2024.
• Kim, J. (2026). Designing Youth Social Media through Problem Space Attunement. DIS Companion 2026.

The case for cross-cultural collaboration

Research on youth social media design, including Kim's own, is overwhelmingly conducted with participants in a single country. That limitation is not incidental. Norms around self-disclosure, relationship initiation, and communication directness vary across cultures in ways that likely interact with design affordances — a platform feature that reads as an invitation in one context can read as an imposition in another. A four-week deployment with 99 young participants across the United States and South Korea observed this proximity without being able to resolve it: the study included both populations, but a systematic cross-cultural comparison lay outside its scope. The same gap appears in the trust study, whose participants were all U.S.-based. Design conclusions drawn from one culture's norms are quietly presented as design conclusions about people.

The Social Media Design Assembly puts that gap in front of students rather than in a limitations section. Each enrolled student joins a standing group of four with university students studying UX design at partner institutions outside the United States.¹ All four members participate as equals; there are no mentor–mentee roles. The Assembly runs two threads from the first week of the quarter.

In the reading exchange, members rotate as discussion lead — preparing a teaching summary and three to five questions, facilitating a mostly asynchronous conversation, and closing it with a synthesis of what the group surfaced. Each member leads twice. In the collective inquiry, the group works through questions no reading settles: what would better social media actually feel like, and how would you know it was working? Which platforms are thriving, and do three countries agree on what thriving means? What were you hoping for when you joined the platforms you use? What are the unwritten rules of posting where you live? The inquiry is cumulative. By mid-quarter, groups are asked to examine the norms they themselves established without ever discussing them — studying online community by having become one.

Both threads live on a moderated Discord server. Discord is a deliberate choice rather than a convenient one: Kim's interview study of 25 Discord users found it is the platform young people most often name as conducive to building real relationships, and identified the design mechanisms responsible — persistent channels that let the same people meet repeatedly, segmentation that regulates social density, visibility cues that make regulars recognizable, tiered engagement that permits lurking before committing. Students inhabit those mechanisms while studying them. In Week 3, having known each other only asynchronously, groups attempt sustained synchronous connection across a rotating set of platforms and document what each affords.

The Assembly is therefore not an enrichment activity attached to the course. It is the course's living case study, and it carries 35% of the course grade. Each group attempts, over ten weeks, precisely what platforms promise and rarely deliver: building genuine community with people they have never met. That attempt supplies the material students reflect on, exposes every participant to how the same interface reads differently across three cultures, and gives external participants a structured occasion to practice the distributed, cross-cultural collaboration that professional UX work demands and rarely trains.

1. Assembly partnerships are currently in discussion with university UX design programs in South Korea and Ireland. Confirmed partner institutions will be named here.

The course

INFO 498C is a four-credit, lab-based studio for advanced undergraduates. It develops two complementary capacities: the literacy to read platform design critically, and the design skill to build alternatives that reflect one's own values. The course turns on a question asked in Week 4: what is social media? Until then, students study the platforms they already know — how features shape behavior, break trust, and answer or fail the interpersonal needs of young people. After the pivot, the frame widens to alternative platforms, games and spatial technologies, speculative design, and the question of whose values any of them embody.

The final three weeks belong to the project. Each student designs and builds a social media platform of their own, then audits it in CodesignStudio, an LLM-agent simulation sandbox that populates a design with synthetic users and reveals how it behaves under realistic social dynamics — including the dynamics its author would rather not think about. The audit is weighted more heavily than the prototype, because it is the component a student cannot fake.

What students leave with

The course is built so that its outputs are legible to employers. Each item below names something a student will have done, in the terms a portfolio, a CV, or an interview answer requires:

• End-to-end product design. Problem framing, user research synthesis, information architecture, interaction design, and a working interactive prototype — a complete case study rather than a screen.
• Heuristic and values-based design critique. A structured teardown of a mainstream platform that traces a specific behavioral outcome to the design decisions that produced it.
• Simulation-based evaluation and safety auditing. Experience stress-testing a design against synthetic users in an LLM-agent sandbox, and reporting what broke — an emerging practice in responsible AI and trust-and-safety work that few graduates have touched.
• Design rationale and documentation. The ability to state which values a design embodies, where it trades them off, and why — the skill portfolio reviewers most often find missing.
• Cross-cultural and distributed collaboration. Ten weeks of asynchronous work with peers on two other continents, in English, across time zones, with a facilitation role held twice.
• Facilitation and design communication. Leading design discussions, teaching technical material to peers, and presenting a defended design position to an audience.
• Applied research literacy. Fluency with current HCI research on youth wellbeing, trust, privacy, and platform governance, digested by teaching it rather than by reading it.

Learning objectives

By the end of the course, students will be able to:

1. Analyze how platform design shapes user behavior, with particular attention to which design patterns support — and which undermine — the developmental needs of youth online.
2. Articulate their own vision of “better” social media and situate it among competing visions, explaining where it aligns with and departs from other people's values.
3. Design and build a working social media prototype that embodies an explicit set of design values.
4. Evaluate their own and others' designs with nuance, surfacing tradeoffs and tensions rather than collapsing them into a single verdict.
5. Collaborate and communicate across cultural contexts, and reflect critically on what that experience reveals about platform design.

Vision and expected contributions

Most social computing research addresses the flaws of existing platforms. That work is necessary, and it is not the whole task: the absence of negative outcomes is not the presence of good, particularly for the identity and relationship work central to young people's development. The Positive Social Technology framework, which Kim co-developed with colleagues across ten institutions, argues for a complementary orientation — moving beyond fixing, protecting, and preventing, toward technologies that support human flourishing. This course is that argument in pedagogical form. Students are not asked to diagnose Instagram. They are asked to build the thing they would rather have, and then to say honestly what it costs.

The course expects to produce three contributions. For students, it produces demonstrable capability rather than exposure: a built platform with a documented simulation audit, an articulated design position situated among competing values, and sustained experience collaborating across cultures under realistic conditions. For the research program, each cohort generates a corpus of platform prototypes and design rationales authored by young people who were given room to define what better means before being asked to build it — an instance of the definitional attunement the dissertation argues for. For the field, the course offers a tested and documented model for embedding international peer collaboration in design education, on a topic where a school's research strengths meet broad public concern.

A quieter ambition sits underneath these. Students in Kim's workshops arrive describing social media's future as doomed and leave describing the exercise as hopeful and agentic. That shift is not decoration on the research; it is a finding about what happens when young people are handed the design problem instead of the warning label. A course that reproduces it at scale, across three countries and ten weeks, would be evidence that the shift generalizes.

Learn more about the course

The full syllabus covers the weekly schedule and readings, assignments and grading, milestone specifications, and course and university policies. It is also available as a PDF.

Prospective Assembly partners, industry critics, and sponsors will find the most relevant details in the questions below, and are welcome to write directly to jaewonk@uw.edu. A detailed partner brief is available on request.

Questions from partners and collaborators