Design Thinking with AI

Eliana Bell and Jayson Danielle Saclo, the Experience Designers on this project, working from the nation's capital and the Mighty Manawatū, have been exploring what it means to carry out Design Thinking when pairing with AI. This article is a collaboration and, we’re going to share what we’ve learned, the challenges we’ve faced, and how this work is reshaping how we think about creating experiences.

Generative AI (GenAI) is undeniably bringing about a significant evolution in how the Design Thinking iterations of Empathise, Define, Ideate, Prototype and Test unfold. While AI excels in areas such as sentiment and data analysis or user story generation, its true power emerges when paired with human expertise and diverse datasets.

• Empathise: Traditionally involving in-depth user research, this phase is now being augmented by AI-powered user research and sentiment analysis. AI can rapidly analyse large datasets to uncover deeper and more accurate insights into user behaviour, automating tasks like interview transcription and integrating learnings from various projects. This allows for quicker and potentially more comprehensive initial understanding. Define: The process of gathering and analysing research to define problem statements is now enhanced by data-driven problem definition and identification. AI's ability to synthesise data and pinpoint key issues enables teams to focus their problem statements more precisely and efficiently.

• Ideate: Brainstorming sessions now benefit from AI algorithms for solution generation. AI can generate a wider range of ideas based on data, suggesting innovative solutions that might not immediately occur to human teams. Designers can then build upon these AI-generated concepts, leading to more diverse and potentially groundbreaking ideas.
• Prototype: The creation of low-cost, experimental product versions is being revolutionised by rapid prototyping powered by AI tools. Platforms like Figma and no-code tools such as Loveable and v0 by Vercel enable designers to quickly create and modify design elements, even generating low-fidelity prototypes rapidly. Iteration in prototyping can now occur at an unprecedented speed.
• Test: Evaluating prototypes with real users is now amplified by utilising AI for feedback-collection and analysis. AI tools can conduct A/B tests and analyse user feedback in real-time, providing immediate data to inform further iterations and refinements. The integration of AI doesn't negate the iterative nature of design thinking; rather, it transforms the shape and speed of these iterations. Instead of the more linear and evenly paced iterations sometimes associated with traditional models, AI enables faster cycles of data collection, analysis, ideation, and prototyping. This could be visualised as a more agile and responsive flow, constantly adapting based on rapid insights – perhaps hinting at the “stingray model” in its dynamic and adaptive nature.

To give us a sense of how this accelerated Design Thinking with AI model would work, here is a quick way using secure and widely available tooling to spin up a prototype for an adjacent Buzzly offering.

However, human oversight remains crucial. AI models need to be calibrated with the right data, and iterative testing requires human judgment to ensure relevance and accuracy. Designers play a vital role in guiding the AI, ensuring ethical considerations are met, and securing stakeholder buy-in for AI-driven outputs.

The evolution of design thinking with AI is not about replacing the fundamental principles but about enhancing its power and agility. By embracing this integration thoughtfully, we can unlock new levels of efficiency, gain deeper user understanding, and ultimately create more impactful and user-centred products. The journey of design thinking continues, now with AI as a powerful co-pilot.

The integration of AI into design thinking fundamentally shifts the designer's responsibilities, moving them beyond purely executing design tasks to becoming strategic orchestrators and critical thinkers. While AI takes on tasks like rapid data analysis, automated prototyping, and initial idea generation, the designer's role becomes more nuanced and crucial in guiding the process and ensuring the final outcome aligns with user needs, business goals, and ethical standards.

The Visionary and Strategist: Designers are no longer solely focused on visual execution. They need to define the problem space, set the strategic direction for AI exploration, and frame the challenges for AI to address. They are responsible for understanding the overarching user needs and business objectives, ensuring that AI-driven insights and solutions are aligned with this broader vision.

The AI Collaborator and Guide: Designers must learn to work collaboratively with AI tools, understanding their capabilities and limitations. This involves providing initial ideas for AI to develop, calibrating AI models with appropriate data, and interpreting the outputs generated by AI. They act as guides, steering the AI towards relevant and user-centric solutions.

The Data Interpreter and Translator: As AI generates vast amounts of data and insights, designers need the skills to interpret this data effectively. This includes understanding the patterns and trends identified by AI and translating these findings into actionable design decisions. Furthermore, they play a crucial role in communicating these data-driven insights to stakeholders in a clear and understandable manner. This often necessitates collaboration with Data Scientists, forming a crucial partnership within the organisation. Designers bring user empathy and contextual understanding, while Data Scientists provide expertise in data analysis and model development. This collaboration ensures that AI-driven solutions are not just technically sound but also user-centred and ethically responsible.

The Ethical Guardian: With the increasing use of AI, designers become central figures in ensuring ethical considerations are addressed throughout the design process. This involves:

Identifying and Mitigating Bias: Designers must be aware that AI systems can inherit biases from the data they are trained on. They need to work with data scientists to ensure diverse and representative datasets are used. Moreover, they play a role in critically evaluating AI outputs for potential biases and ensuring fairness and inclusivity in the final product. Regular human audits are essential in this process.

Ensuring Transparency and Explainability: Designers need to advocate for transparency in how AI is used and ensure that users are informed about the AI's role in the product or service. They may also need to work with data scientists to understand and explain the logic behind AI-driven decisions, fostering user trust.

Obtaining User Consent: Designers must ensure that ethical and legal requirements regarding data privacy are met, including obtaining explicit user consent for data collection and analysis.

The Human Intelligence & Oversight Advocate: Despite AI's capabilities, human oversight remains indispensable throughout the AI-augmented design thinking process. This oversight is crucial in several stages:

Defining the Problem: While AI can analyse data to identify potential problems, human designers are needed to frame these problems in a user-centric and strategic way.

Calibrating AI Models: Ensuring that AI models are trained on data that accurately reflects the target audience and market requires human expertise and judgment.

Evaluating AI-Generated Ideas and Prototypes: Designers must critically evaluate the feasibility, desirability, and viability of AI-generated solutions, ensuring they align with user needs and business goals.

Conducting User Testing and Incorporating Feedback: While AI can assist with feedback analysis, direct interaction with users and the interpretation of qualitative feedback remain crucial human tasks. Designers ensure that human empathy informs the iterative refinement of solutions.

Securing Stakeholder Buy-in: Presenting and justifying AI-driven design decisions to stakeholders often requires human communication skills, persuasion, and the ability to build trust. Workshops and stakeholder involvement in prototyping sessions are vital for this.

In essence, the designer's role evolves from a pixel-pusher to a strategic leader, a collaborative partner with AI and data scientists, an ethical guardian, and a champion for human-centred outcomes. Their ability to understand how AI finds solutions, to collaborate effectively with data scientists, and to maintain a strong focus on ethics and human oversight will be paramount to successfully navigating the future of design. Whilst we can’t predict the future, we can help shape it! - Jayson Danielle Saclo, & Eliana Bell, Experience Designers.