Ensuring AI Trust and Transparency with Inclusive Design

Editor's Notes

• #1: Ensuring AI Trust and Transparency with Inclusive Design We must balance the value of AI branding with the customers' need to understand their risk of accepting Gen AI content, providing data, and understanding the AI process. Explore trauma-informed principles and inclusive design solutions.
• #3: Historically, accessibility specialists focused on a narrow set of disabilities. We focused on the senses, such as sight, sound, and touch. We focused on abilities, like hearing, movement, and seeing. We expanded to include cognitive, mental health, and neurodiversity. This is significant. We now have tools to build inclusive products and services for an estimated 25% of the population. What about the other 75%? As accessibility professionals, we understand unique experiences and needs. We are best equipped to expand customer research and design at our companies. Universal design was described as a one size fits all solution. Inclusive design is one size fits one. Intersectional design is one size fits one, but also accounts for price, texture, availability, cultural appropriateness, and more. This presentation introduces the next layer of inclusive design; one that recognizes trauma. Trauma-Informed Design (TID) started in education, health, and community spaces. It focuses on the person’s experiences, recognizing trauma’s impact, anxiety, and restoring personal control. Architects embraced TID to develop spaces that are comfortable instead of confrontive. While the earlier stages of TID focused on individualized experiences, we can still take the principles and apply them to web and mobile application design. This is especially critical for emerging AI powered experiences where transparency and collective understanding are rarely considered. Slides: https://www.slideshare.net/slideshow/introduce-traumainformed-design-to-your-organization-csun-stc-2024/266889804
• #4: Trauma-informed design (TID) principles were established in architecture and service design with an emphasis on providing safe spaces for those who have suffered trauma. More recently, TID has been adopted within digital design of web and mobile applications. The core principles of TID also lead to usable design that respects the user’s privacy, security, and trust. Safety Trustworthiness and Transparency Peer Support Collaboration and Mutuality Empowerment, Voice, and Choice Cultural Consideration We are also at an inflection point with the integration of Artificial Intelligence within our applications and services. Incorporating the TID principles in AI-influenced design will ensure our products are supporting our diverse audiences and prioritizing their experience. 5 Principles of Trauma-Informed Design Trauma-informed Design Resources
• #5: The most important element of TID is to do no harm. Our work will focus on ensuring the customer understands the risk of generative AI content, recognizes when AI is being used to generate information, understands how the decision was generated, and the ability to exit the experience when there is a concern. We will take an accessibility-first approach, focusing on the experience of assistive technology users and cognitive disabilities. Building upon this foundation will ensure our AI interfaces work well for everyone. Threats by artificial intelligence to human health and human existence - PMC
• #6: Some AI is low risk. For instance, prompting generative AI to create a cartoon of a cat skateboarding is probably not going to hurt the customer. Asking AI to transform a sketch into an estimate/invoice and send to a customer has multiple opportunities for error and has a high impact on the business and customer . A customer should understand generated content and actions can include errors and must be reviewed. The process of generating an answer must also be transparent.
• #7: To do no harm, we must be transparent about the role of AI and the potential risk of accepting content without a review. AI identification needs to be clear and accessible. High risk pathways must confirm approval. Describe how a decision was made via Explainable AI (XAI). Provide multiple options. Resources Screened Out: The Impact of Digitized Hiring Assessments on Disabled Workers - Center for Democracy and Tech Confirmation Dialogs Can Prevent User Errors (If Not Overused) What is Explainable AI (XAI)? | IBM UX considerations for generative AI apps and agents | Google Cloud Blog
• #9: Clearly identifying AI Many products use a combination of sparkles, propellers, and rainbow borders to highlight information is AI generated. AI UX Patterns | Symbols | ShapeofAI.com
• #10: In this example, the alt text on the icon was “sparkle”. This doesn’t tell the customer that AI was used to identify 10 outstanding invoices and providing additional payment options. Change the to “AI powered”. I’ve asked several screen reader users about using “AI” in alt text. It’s a common acronym at this point, you don’t need to do anything to separate the letters. Video: https://youtu.be/C-rmJhnSyVE
• #11: At Intuit, we treat SVG icons as decorative via the aria-hidden rule. If they provide context, they are wrapped in a container, such as a button, link, or span that includes what the icon represents. Bad (SVG) <title>Thinking and generating</title> <desc>Brand animation for thinking and generating state</desc> Good <div role="img" aria-label="AI Powered"> <svg aria-hidden="true"> </svg> </div> Video: https://youtu.be/3KItu7RoGbM
• #12: Rainbow borders and animations are often used for defining something that is AI powered. This is fine if there is also content to describe the AI connection. This also includes the use of AI brand names, such as Rufus, Ari, and Gemini. In chat interfaces, it’s important to label the responder as an AI agent, Amazon’s Rufus uses an image with alt=””. <img alt="" src=".." class="rufus-avatar-img" height="20" width="20">
• #13: Explainable AI In many ways, Explainable AI (XAI) is the antidote for irresponsible, harmful bias in AI. Imagine you are fired from your job after many years of solid work, only to find out the decision was made by an AI service. If the decision used XAI, the process would be clearly defined and open for review.
• #14: Cash flow is critical for small business success. QuickBooks introduced an invoice reminder service that highlights past due invoices and suggests emails to send to customers. This includes several options for the voice used, such as fun or firm. We will use this screen to illustrate several best practices.
• #15: There’s also a clear explanation of why this invoice was highlighted and the customer’s history. This is used to define the style of letter being sent.
• #16: Customer Choice Choice and control are critical for designing with a trauma-informed approach. Our goal is to reduce anxiety and ensure the customer has control of their interactions. Continuing with the QuickBooks example, there are several sections that ensure choice and control. Voice and Tone chooser While there are some guidelines for choosing the tone of a reminder email. A friendly message is great for an invoice that is only past-due a few days. But a stern message is needed for those that are overdue by more than 30 days. The customer is always able to choose what they feel comfortable with.
• #17: Exit the AI Experience There’s a button to exit the generative AI functionality and send a basic past-due invoice reminder. The customer can still send an email, but trust the content is not AI generated. Safe exit is critical for trauma-informed design, especially when using an application could lead to physical or mental harm. For example, an application for reporting human trafficking.
• #18: Editable text Each of the email attributes can be edited. This allows the customer to review the generated content to spell check, correct contact information, and add personal notes.
• #20: You own a bakery in Anaheim and there will be a new Mosque built in the community that supports the Iranian community. You want to see how you could add recipes to welcome the new residents and generate sales. You have agentic ai review your past sales and recipes for the use of rose water. You want to compare with other nearby bakeries for rose-based desserts. You want to analyze the number of bakeries near the mosque. Finally, create a business plan for increasing the selection of deserts that include rose water for the next five years. The majority of this work will not have an interface, it’s the collection of data from multiple sources. As inclusive designers, we need to make sure the resulting data can be verified, is understandable, and the customer has options to expand on the information.
• #21: Integrating trauma-informed design (TID) principles into AI design is crucial for building trust and ensuring inclusivity. By prioritizing safety, transparency, and user choice, we can create AI experiences that are not only functional but also respectful and considerate of diverse user needs and potential vulnerabilities. As AI continues to evolve, a commitment to these principles will be essential for mitigating risks, preventing harm, and fostering a positive and accessible experience for all users.