Sometimes, default AI does not feel like your AI. The way it speaks, with metaphors, quick topic shifts, or hidden assumptions, can trip up neurodiverse people who thrive on clarity and predictability. A small tweak using custom instructions can transform it into an empathetic ally instead of a chaotic chatterbox.
Why neurodiverse users sometimes clash with default ChatGPT
• Language with implied meaning and figurative expressions may not land as intended.
• Rapid shifts in topic combined with inconsistent formatting or tone can create confusion.
• Long, digressive answers or overly short ones that need follow-up questions both overload working memory.
Source: Access-Ability Newsletter – https://buttondown.com/access-ability/archive/adapting-chatgpt-to-meet-neurodivergent/
The beauty of custom instructions
Setting these once is like planting markers in a maze. You tell ChatGPT how you want to communicate:
• Define tone, formatting, or level of detail.
• Make responses structured with headings, bullets, and consistent order.
• Avoid figurative language unless you specifically want it.
• Ask ChatGPT to say “I don’t know” instead of avoiding the answer.
• Include a “STOP” signal so you can halt long answers when you need to.
Source: Access-Ability Newsletter – https://buttondown.com/access-ability/archive/adapting-chatgpt-to-meet-neurodivergent/
Why it matters and what it shows
Personalizing your AI is not about making it coddle or dumb itself down. It is about reshaping the interface so that it meets your brain’s unique wiring. A voice that is logical, a layout that is predictable, and a rhythm that does not scatter ideas.
This is not just a convenience. It is accessibility in practice. Instead of forcing everyone into a one-size-fits-none format, you can tailor the tech to your needs. That is how inclusion thrives, by valuing individual ways of thinking instead of forcing conformity.
Source: Access-Ability Newsletter – https://buttondown.com/access-ability/archive/adapting-chatgpt-to-meet-neurodivergent/
Final thought
A tiny tweak in how ChatGPT communicates to match your style is not just neat. It is empowering. It turns AI into something collaborative, predictable, and aligned with your flow. If inclusive design were a conversation, this is it, where your voiced preferences shape how the tech listens back.
If generative AI becomes part of daily life for most people, it will work best when its output is shaped to meet individual needs. A single “default” communication style will never serve everyone equally.
https://buttondown.com/access-ability/archive/adapting-chatgpt-to-meet-neurodivergent/
unknownx500





