Pattern 47
Visible work and recoverable failure
Describe real progress, responsibilities and outcomes.
Concrete takeaway
Show meaningful steps, their actual state and who needs to act next.
Choose an approach
These are editorial decision conditions to validate. Some alternatives are successive states or can be combined; they are not always exclusive choices.
| Alternative | Use when | Tradeoff or requirement |
|---|---|---|
| A · Named steps | Work involves several named steps | A progress display must reflect real work |
| B · Review responsibility | Responsibility is divided between AI and people | The assigned reviewer must actually exist |
| C · Partial failure | Only part of a task succeeded | Useful results and unresolved gaps must remain separate |
The problem
An agent can appear busy without revealing whether it retrieved sources, created a draft or changed saved data.
The pattern
Show meaningful steps, their actual state and who needs to act next. Use concise process explanations rather than presenting hidden model reasoning as a faithful transcript. On failure, name what succeeded and offer a safe retry or alternate route.
Compare the alternatives
Describe real progress, responsibilities and outcomes.
A · Named steps
- Retrieve selected sources
- Prepare a draft
- Wait for your review
Progress describes actual work.
B · Review responsibility
The next decision belongs to the named actor.
C · Partial failure
Successful work is preserved without duplicating actions.
Original wireframe proposals · No live controls · Grey rows represent schematic text, not loading states · Labelled placeholders are not Scripture quotations
Download this wireframe as SVGFailure modes
Fake progress percentages; “done” before saving; hiding partial changes; confident claims of a human review that never happened.
What to validate
Test partial retrieval, failed saving and permission refusal. Can users identify the completed work, pending action and safe recovery path?
Research-informed design proposal. These visual alternatives have not been tested with users in this atlas. Source findings, documented behaviour and this proposed adaptation are different kinds of evidence.
Sources and adaptation
This card is an original design synthesis. The following sources inform its content distinctions, interaction approach or review requirements; they do not validate the whole pattern.
- AI Interaction AtlasOfficial repository and taxonomy reviewed · A taxonomy, not a prescriptive UI framework. The official repository is available at https://github.com/quietloudlab/ai-interaction-atlas under Apache 2.0.
- Agentic UX PatternsPublished catalogue overview and example reviewed · The catalogue presents 49 proposals. Their inclusion is not evidence of effectiveness in faith or clinical settings.
- AI UX PlaygroundCatalogue and selected detail pages reviewed · Selected relevant entries reviewed; the entire catalogue was not audited. Its FAQ currently describes 150 patterns in 11 categories.
- AI and Design SystemsPublic course description reviewed · Paid lessons and community discussions were not reviewed. Course claims are not research evidence.
Common scenarios and flows
- Prepare and participate in an AI-assisted practice — Use bounded adaptation while keeping Scripture, authorship and user choice inspectable.
- Listen to an AI overview and inspect a claim — Use explanatory audio while retaining access to the source material and control of the session.
Design principles in this decision
Editorial application of Missional by Design. These values frame review questions; they do not validate a pattern’s effectiveness.
- Honesty — Could this interface lead a person to infer more authority, certainty, privacy or human support than the service provides?
Compare alternatives using the task and evidence above. Record competing needs instead of treating a principle as an automatic verdict.