Embedded Chatbots and Accessibility Compliance

 

Table of Contents

General WCAG Chatbot Guidelines 

While there is currently no comprehensive and agreed-upon method for evaluating or recommending accessibility for chatbots, here are some general WCAG guidelines when embedding them to your website content. 

Chatbots can work for many people, and particularly for people with cognitive disabilities if they:

  • recognize misspelled words,

  • provide human contact related information if the chatbot is unable to provide a satisfactory response after 3 attempts, and

  • can be dismissed with a single interaction, and recalled using a link or button.

This is covered in more detail under the WCAG 3.2.6 Level A guideline to provide a consistent help mechanism on a website.

https://www.w3.org/WAI/WCAG22/Understanding/consistent-help#examples  

Also, embedded chatbot elements must integrate with the larger website experience. 

Chatbot Benefits for Users with Cognitive Disabilities 

Enabling users (especially those with cognitive disabilities) to find solutions while communicating their question using their own words (for example by interacting with a chatbot) could increase their chances of success for completing a task. 

Chatbot Accessibility Challenges Based on User Research/Testing 

Chatbots have interface features and interactions that are not fully addressed by WCAG.

Here are a few that should be taken into consideration. 

Message Length, Pacing and History 

Since chatbots deliver messages in sequence, they raise questions about message content length and pacing for users with different types of needs. 

Other issues include how the user should be alerted to new messages as they arrive. For each message, depending on whether the chatbot supports free-text input and/or selection from a set of options, keyboard focus must be positioned relative to web elements to align with user expectations. 

Over time, responses accumulate in a lengthy “conversation history,” which the user should be able to access and review. For example, what are the most accessible ways to present that history and make it easy for user to navigate (especially for users who rely on using assistive technology)? 

Communication Style 

Since chatbots interact conversationally, their communication styles could affect users. For instance, chatbots exhibiting empathy can improve human mood. This could introduce some challenges to ensure chatbots do not negatively impact users emotionally or psychologically. 

Third Party Chatbot Platforms 

Third-party chatbot development platforms might only allow for limited adjustments. This means that accessibility needs to be considered when selecting a platform and when customizing to meet user needs.

Chatbot Accessibility Playbook 

Here is a link to a Chatbot Accessibility Playbook with recommendations for integrating accessible chatbots and how these recommendations map to specific WCAG and COGA  (making content usable for people with cognitive and learning disabilities) guidelines.   

https://mitre.github.io/chatbot-accessibility-playbook/docs/4_2_22.html  

This Playbook was developed by MITRE’s Independent Research and Development Program with the intent that it will be incorporated to the WCAG guidelines.