““Success in creating AI would be the biggest event in human history. Unfortunately, it might also be the last, unless we learn how to avoid the risks.”
-Stephen Hawking”
AI can generate inaccurate information and cannot cite sources reliably
Understanding the Risks
Understanding how AI affects privacy, security, and bias are paramount.
I do not put sensitive data into an AI, unless instructed.
AI-generated designs may prioritize efficiency over emotional intelligence. Leading them to be impersonal, frustrating, or difficult to use. So not a user-centered designer.
Some tasks like documentation and brainstorming are a good fit, while solving a specific design problem may take longer and give less effective results.
Azi and (Robot) Levi from Scavengers Reign. Levi is there to help but also puts Azi in danger.
Playing the Field
Experimenting is useful to discover the benefits and limitations of new tools/versions.
Comparing results/UI’s between different tools helps me gauge their strengths and weaknesses.
Currently, I prefer Claude’s UX due to it’s inviting tone and nature look and feel.
Currently bouncing between Claude, Gemini, ChatGPT, and Figma AI,
Prompt Research
It’s so fast at analyzing large amounts of secondary research, analyzing top competitors, and generating potential user personas.
I would not solely rely on since there’s a risk of inaccurate information.
I use the five step framework from Google’s AI Prompt Engineering course: Task, Context, References, Evaluate, and Iterate.
Using different prompts/tools to accelerate my research
Collaborating Tools
While in discovery for a new feature or product, I will lead discovery sessions in FigJam with their AI Jambot.
It’s a simple tool that can create some wide reaching discussions to get the team approaching the problem together.
I’ve found it adds a lot of engagement and can decrease the scope of projects.
Jambot in FigJam provides a wide range of tools to elevate team brainstorming
“As it stands, it’s highly unlikely that AI will replace the need for UX designers. Ultimately, UX is far too reliant on the ‘human touch’; empathy will always be number one when it comes to designing user-friendly products.”
Human Design, AI Helper
I’ve experimented with the suite of Figma’s AI tools. There are a few that I have incorporated into my workflow.
‘Add Interactions’ can be useful when attempting new or time consuming prototype interactions.
‘Rename layers’ is useful when I’ve neglected to name my layers and am sharing the design file.
‘First Draft’ is helpful to see how it handles the components and hierarchy of a new feature.
Figma’s built-in suite of AI tools for designing
Looking Forward
As of now, I find designing manually is still faster and more reliable for complex problems. However, I continue to explore and experiment with AI in my workflow as it can speed up discovery and iteration time to deliver products that help people and businesses achieve their goals.
