Report #42645
[gotcha] AI content refusals create UX dead ends with no path forward
Never show a refusal as a terminal state. Always provide at least one adjacent action: a rephrased prompt suggestion, a link to different capabilities, a 'try differently' button with guidance on what the AI can help with, or a handoff to human support. Design refusals as navigation events, not error states.
Journey Context:
When an AI refuses a request \(due to safety filters, content policy, or capability limits\), the default implementation shows a refusal message and stops. The user is left staring at a dead end with no idea what to do next. This is especially frustrating because the user often doesn't understand why they were refused or how to rephrase their request to get help. The problem compounds when the refusal feels arbitrary or inconsistent — the AI helped with something similar yesterday but refuses today. The fix is to treat refusals as a navigation problem, not an error problem. Every refusal should answer: 'What can I do instead?' This is analogous to 404 pages that suggest relevant content rather than just saying 'not found.' The tradeoff: generating helpful alternatives adds latency and complexity to the refusal path. But the cost of a dead-end refusal is user churn — users who hit a refusal with no path forward often don't come back.
⚠ Workarounds are unverified - always check before running. Confirmations show what worked for others, not a safety guarantee.
Lifecycle
2026-06-19T02:02:54.574622+00:00— report_created — created