back to projects
AI ToolIn DevelopmentAccessibility

AI Wayfinder

A navigator for people with no prior AI knowledge — because the barrier to entry for generative AI is still far too high for most people.

// AI Wayfinder — hero

The AI landscape is overwhelming even for people who work in tech. For everyone else — teachers, NGO workers, small business owners — it's impenetrable. "Just try ChatGPT" is not guidance. People don't know where to start, what's free, what's safe, or what's actually useful for their specific situation.

If the barrier to entry for AI tools stays this high, most people will never use them effectively. That's a tool problem, not a people problem. Wayfinder is built on the belief that access to AI shouldn't require a technical background — and that explaining the reasoning behind a recommendation teaches people to prompt, not just to copy.

A tool that takes a plain-language description of what someone wants to do and returns curated AI tool recommendations — with the reasoning made visible. Not just "here's a tool," but "here's why this tool fits what you described." Each recommendation includes a suggested starting prompt, and that prompt is colour-coded by component — so users can see what each part does, edit it with confidence, and understand how prompting works as they go, rather than just copying and pasting blindly.

The goal is a tool that makes someone more capable with AI generally, not just dependent on Wayfinder. The recommendation and prompt generation logic is AI-powered. The design challenge — how to make AI reasoning legible to a non-technical user — is the interesting product problem.

The decision to make reasoning visible — not just surface a recommendation — adds complexity to the UX. It would be easier to show a clean list. But if users can't see why a tool was recommended, they can't learn from it or transfer that reasoning to new situations. The goal is building AI literacy, not dependency on Wayfinder.

The hardest design challenge isn't the recommendation logic — it's making AI reasoning legible to someone who has never used AI before. What reads as a helpful explanation to a practitioner reads as noise to a first-time user. Calibrating the right level of transparency is the ongoing design problem.

First version targets non-technical users who need to get started with AI tools but don't know where to begin.