Designing for the Next Billion Users in AI
Research
Product
Designing for the Next Billion Users in AI
How we're approaching accessibility and inclusive design as artificial intelligence becomes the primary interface for work across the globe.
Query builders, filter panels, and prompt boxes all share an assumption: that the person using them can already articulate what they are looking for. For the next wave of users, that assumption does not hold. The people who most need decision intelligence are often the ones least equipped to interrogate a database — and they are the majority of the market, not the edge case.
Start from the decision, not the data Interfaces organised around business questions outperform interfaces organised around tables, because the user recognises the question even when they cannot construct the query. Make the system explain itself Every answer includes what it looked at and how confident it is. Trust is built by showing work, not by hiding it behind a confident tone. Assume low bandwidth and older devices Our target users are frequently on constrained connections. Performance is an accessibility requirement, not a polish item. Support more than one language properly Translation is the floor. Number formats, date conventions and reading direction are where most products quietly fail.
We stopped tracking feature adoption as a primary metric and started tracking how many users reached a decision without contacting support. It is a harder number to move and a far more honest one.
"Accessibility isn't a mode you switch on. It's whether the product works for the person who wasn't in the room when you designed it."
No spam. Unsubscribe anytime.