As technology advances at breakneck speed, UX design frameworks struggle to keep up. The emergence of Generative AI (GenAI) systems — capable of creating novel text, images, and even software solutions — has redefined the role of design in digital products. Yet, the methods we rely on to design these experiences often feel like trying to fit a square peg into a round hole. Conventional design frameworks simply weren't built to handle the complex, dynamic nature of GenAI systems.

The UX Design Gap

Traditional UX frameworks, from Double Diamond models to design thinking playbooks, thrive in environments where user interactions follow predictable paths. They guide teams through static workflows, with clear user goals and defined error states. But GenAI isn't like that. It's messy, iterative, and full of uncertainties. The output isn't always predictable, nor is the way users interact with it. Imagine designing for an AI system that writes poetry based on a user's mood. What happens when the user isn't happy with the poem? How do we handle biases or ethical dilemmas in generated content? These challenges require design frameworks that embrace uncertainty, support dynamic user flows, and incorporate real-time feedback.

Challenges Faced by Designers and Practitioners

Designers, developers, and product managers working with GenAI face countless hurdles:

• Error handling: How do you design interfaces that gracefully manage unexpected or nonsensical AI outputs?
• Ethics and trust: How do you ensure users trust the AI when its outputs can be opaque or even controversial?
• Iterative design: How can teams test and refine AI-driven experiences in iterative cycles when outcomes are unpredictable?
• Variable interfaces: Unlike static apps, GenAI systems require interfaces that adapt to different user needs and states.
• Existing tools like design cards and UI kits often fall short because they weren't built with these complexities in mind.

My Current Research

These challenges and gaps are at the core of my current research, which focuses on exploring how we can develop practical, actionable design frameworks tailored for GenAI applications. I’m investigating the limitations of existing design resources and seeking to understand how GenAI practitioners engage with these frameworks — or why they often don’t. Ultimately, the goal is to propose methods and tools that make design frameworks more relevant, adaptable, and useful for real-world GenAI systems.

A Call for Practical, Actionable Frameworks

If we're going to design meaningful, user-centered GenAI experiences, we need a fundamental shift in how we approach UX design. We must go beyond abstract guidelines and develop practical, adaptable resources that account for the unique challenges of GenAI. This means:

• Creating iterative design processes that support rapid testing and feedback loops.
• Developing new patterns and tools to address variable interface states, error management, and user trust.
• Integrating ethical considerations at every stage of the design process.
• Building customizable UI kits that prioritize user flows rather than static components.

Bridging the Gap Between Theory and Practice

Academic research has started to identify high-level principles for GenAI design, but these principles often remain theoretical. What’s missing are actionable methods that practitioners can easily integrate into their workflows. Companies like Google and Microsoft have developed AI design guidelines, but they aren't always adaptable to the dynamic needs of GenAI systems. There's a clear opportunity to bridge this gap by developing more practical and context-sensitive design resources.

Looking Ahead

As GenAI continues to evolve, UX designers, developers, and product managers have a crucial role to play. By rethinking and adapting our design frameworks, we can ensure that GenAI applications are user-centered, transparent, accessible, and ethical. It's an exciting time to be in design, but we need to rise to the challenge and build the tools and processes that will enable us to navigate this new frontier. Through my research, I hope to contribute to this ongoing effort — creating a foundation for UX design that keeps pace with the ever-changing landscape of GenAI applications. Like all meaningful design work, this evolution will be iterative, and I'm excited to be part of the journey.

References

AMERSHI, Saleema et al. 2019. ‘Guidelines for Human-AI Interaction’. In Proceedings of the 2019 CHI Conference on Human Factors in Computing Systems. CHI ’19: CHI Conference on Human Factors in Computing Systems, Glasgow Scotland Uk, 2 May 2019. Available at: https://dl.acm.org/doi/10.1145/3290605.3300233. BALASUBRAMANIAM, Nagadivya, Marjo KAUPPINEN, Sari KUJALA and Kari HIEKKANEN. 2020. ‘Ethical Guidelines for Solving Ethical Issues and Developing AI Systems’. In Product-Focused Software Process Improvement. Cham: Springer International Publishing, 331–46. Available at: https://dl.acm.org/doi/abs/10.1007/978-3-030-64148-1_21. CLARK, Colin. 2024. ‘Design Handbook’. [online]. Available at: https://fluidproject.atlassian.net/wiki/spaces/fluid/pages/11540909/Design+Handbook [accessed 22 Dec 2024]. DENG, Wesley Hanwen et al. 2022. ‘Exploring How Machine Learning Practitioners (try To) Use Fairness Toolkits’. arXiv [cs.HC]. Available at: http://arxiv.org/abs/2205.06922. ‘Design Patterns for Fluid Navigation – How to Use Inline Linking’. n.d. [online]. Available at: https://www.interaction-design.org/literature/article/design-patterns-for-fluid-navigation-how- to-use-inline-linking?srsltid=AfmBOorQPPwk9nuKUiKxLyye1FD_4-NHsJDHeY-i5RDJyJXH NdlXt_ta. FUKUDA-PARR, Sakiko and Elizabeth GIBBONS. 2021. ‘Emerging Consensus on “ethical AI”: Human Rights Critique of Stakeholder Guidelines’. Global policy 12(S6), [online], 32–44. Available at: https://onlinelibrary.wiley.com/doi/abs/10.1111/1758-5899.12965. HAGENDORFF, Thilo. 2020. ‘The Ethics of AI Ethics: An Evaluation of Guidelines’. Minds and machines 30(1), [online], 99–120. Available at: https://link.springer.com/article/10.1007/S11023-020-09517-8. IDEO.ORG. 2014. ‘How Might We’. [online]. Available at: https://vimeo.com/102964749 [accessed 30 Apr 2024]. KIM, Neunghoe. 2020. ‘User Experience Validation Using the Honeycomb Model in the Requirements Development Stage’. The International Journal of Advanced Smart Convergence 9, [online], 227–31. Available at: https://koreascience.kr/article/JAKO202028959353177.page. LI, Tianyi, Mihaela VORVOREANU, Derek DEBELLIS and Saleema AMERSHI. 2023. ‘Assessing Human-AI Interaction Early through Factorial Surveys: A Study on the Guidelines for Human-AI Interaction’. ACM transactions on computer-human interaction: a publication of the Association for Computing Machinery 30(5), [online], 1–45. Available at: https://www.researchgate.net/publication/359958842_Assessing_Human-AI_Interaction_E arly_through_Factorial_Surveys_A_Study_on_the_Guidelines_for_Human-AI_Interaction. ‘Machine Learning’. 2024. Apple Developer [online]. Available at: https://developer.apple.com/design/human-interface-guidelines/machine-learning [accessed 22 Dec 2024]. MADAIO, Michael A., Luke STARK, Jennifer WORTMAN VAUGHAN and Hanna WALLACH. 2020. ‘Co-Designing Checklists to Understand Organizational Challenges and Opportunities around Fairness in AI’. In Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems. CHI ’20: CHI Conference on Human Factors in Computing Systems, Honolulu HI USA, 21 April 2020. Available at: https://dl.acm.org/doi/10.1145/3313831.3376445. NIELSEN, Jakob. 2023. ‘AI: First New UI Paradigm in 60 Years’. Nielsen Norman Group [online]. Available at: https://www.nngroup.com/articles/ai-paradigm/ [accessed 23 Oct 2024]. NIELSEN, Jakob and Rolf MOLICH. 1990. ‘Heuristic Evaluation of User Interfaces’. In Proceedings of the SIGCHI Conference on Human Factors in Computing Systems Empowering People - CHI ’90. the SIGCHI conference, Seattle, Washington, United States, 1990, 249–56. Available at: https://scholar.google.com/citations?view_op=view_citation&hl=en&citation_for_view=TR6r PkkAAAAJ:u5HHmVD_uO8C. ‘People + AI Guidebook’. 2024. [online]. Available at: https://pair.withgoogle.com/guidebook/ 12 [accessed 17 Dec 2024]. ROWE, Peter G. 1991. Design Thinking. London, England: MIT Press. Available at: https://www.google.com/books?hl=en&lr=&id=ZjZ3mflzJtUC&oi=fnd&pg=PA1&dq=design+t hinking&ots=K96atV6uI1&sig=HBk1jsGLP0NH8yWn9jcq3LhY6Dk. SAVANNAH KUNOVSKY, EVAN BURTON, ESHA REDDY, STEPHANIE SIZEMORE. 2024. ‘A Legal and Ethical Framework For Gen AI’. [online]. Available at: https://www.ideo.com/journal/a-legal-and-ethical-framework-for-gen-ai [accessed 22 Dec 2024]. ‘The Double Diamond’. n.d. Design Council [online]. Available at: https://www.designcouncil.org.uk/our-resources/the-double-diamond/. VU, K. P. L. and R. W. PROCTOR. 2011. Handbook of Human Factors in Web Design, Second Edition. 2nd edn. Edited by Kim-Phuong L. Vu and Robert W. Proctor. Boca Raton, FL: CRC Press. Available at: https://books.google.com/books?hl=en&lr=&id=ZVXRBQAAQBAJ&oi=fnd&pg=PR2&dq=C %C3%A9line+Mariage,+Jean+Vanderdonckt,+and+Costin+Pribeanu.+2011.+State+of+the +Art+of+Web+Usability+Guidelines.+In+Handbook+ofhuman+factors+in+Web+design,+Ki m-Phuong+L+Vu+and+Robert+W+Proctor+(Eds.).+Crc+Press.&ots=q_4o-DDMow&sig=ZIr YRYjBTccosfdtK_o7o5PDEQg. WEISZ, Justin D. et al. 2024. ‘Design Principles for Generative AI Applications’. In Proceedings of the CHI Conference on Human Factors in Computing Systems. CHI ’24: CHI Conference on Human Factors in Computing Systems, Honolulu HI USA, 11 May 2024, 1–22. Available at: https://dl.acm.org/doi/abs/10.1145/3613904.3642466. ‘What Is Generative AI?’ n.d. [online]. Available at: https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-generative-ai. As a list of figures: MED, Badr Chemmaoui. 2018. UI Wireframe Saturday. Unsplash [online]. Available at: https://unsplash.com/photos/blue-ballpoint-pen-on-white-notebook-ZSPBhokqDMc [accessed 28 Jan 2025].