Generative AI is reshaping our mental models of how products work. Product teams must adjust.

🌈 Abstract

The article discusses how generative AI is reshaping people's mental models of how products work, and how product teams must adjust to this change. It explores the concept of mental models, their evolution with the emergence of generative AI, and provides practical guidance for UX designers on how to consider mental models when designing AI products.

🙋 Q&A

[01] Mental Models and Generative AI

1. How do the authors define the term "mental model"?

• Soojin defines mental models as internal representations that people draw on to explain how something works, which can be formally taught or result from shared wisdom and individual experience.
• Mahima defines mental models as people's belief systems that help set expectations for what a product can and can't do, how they can interact with it, the degree to which they can trust it, and what kind of value they can expect to get from it.

2. Why is it important for UX designers to consider mental models when designing AI products?

• Mismatched mental models can lead to unmet expectations, frustration, lack of transparency, misuse, and product abandonment, which can erode user trust.
• Considering mental models can help UX designers adjust and shape the generative AI user experience in products.

3. How are people's emerging mental models of generative AI products different from non-generative AI mental models?

• With generative AI, people interact directly with the AI models, which can be ambiguous and abstract, and culturally insensitive, unlike pre-scripted product interfaces.
• People may anthropomorphize generative AI and ascribe qualities of self-awareness, leading to potential misunderstandings.
• There is a need to shift people's mindset from seeing AI as a purpose-built tool to seeing it as a collaborator.

[02] Practical Guidance for UX Designers

1. What are some practical ways for UX designers to help users update their mental models when interacting with generative AI?

• Identify opportunities and moments to help users calibrate their mental models, especially when interacting with human-like forms that don't look or sound natural.
• Use contextually-appropriate explanations and interactions that factor in people's existing knowledge and mental models of the AI product.
• Help users build trust in generative AI as a partner, not just a tool, by creating moments where AI can seek permission "to know" (user intent) and permission "to act" with those goals in mind.
• Over time, focus on creating AI that collaborates with users (co-shaping and re-finding the results together), rather than merely executing tasks.