AI capability boundaries, agent control, source grounding, uncertainty, approval, and safe fallback.
Browse all problem familiesAn automated system should expose what it is doing, the evidence behind its output, the limits of its authority, and the action a person can take next.
Strong: An assistant drafts a reply with cited policy passages, shows the pending action, and asks a staff member to review before sending.
Weak: An assistant sends a confident reply from incomplete policy context and leaves staff to discover the action through a later customer complaint.
Try it: Map one automated action from request to evidence, proposal, review, execution, and correction.
Begin with the core behavior, apply it in different contexts, then review cases that reveal the quality boundary. Open each entry for its decision rule, states, example, and source trail.
0 of 23 studies complete
Start this family →Learn the shared task and the core interaction choices.
An existing AI feature changes enough that users need to recalibrate expectations before relying on it again.
Users must write or revise an AI request before generation, analysis, transformation, or automation begins.
The AI surface supports open-ended requests and users need examples of useful, supported tasks.
A user's AI request has missing object, audience, timeframe, source, workspace, permission, output-depth, or action-target boundaries.
Users need answer-wide evidence coverage before trusting generated content.
Users must judge whether an AI prediction, classification, recommendation, extraction, risk score, or generated answer is reliable enough to use.
An AI agent, workflow, deployment, or automation is ready to perform a high-impact step and must pause for human authorization.
Users can identify wrong, unsupported, stale, unsafe, biased, irrelevant, or wrongly sourced AI output after it is generated.
Choose a specialized form for the context and follow its complete behavior.
The user needs a back-and-forth assistant conversation with follow-up questions and answer refinement.
Generated text or structured content can be read or monitored before completion.
Users need to verify generated claims, summaries, recommendations, or extracted facts against source material.
Generated content is expected to be revised before it is copied, saved, sent, published, or applied.
A user needs another AI-generated answer for the same request or a visible recovery path after response failure.
An AI agent or automation calls tools, functions, APIs, retrieval systems, commands, or integrations.
An AI agent or automation can show a proposed multi-step plan before execution.
An agent or automation run has started and spans multiple steps, tools, gates, or side effects.
Users need to create or edit automations that run later based on events, conditions, schedules, or record changes.
A generated AI output can influence compliance, customer communication, security, legal, finance, operations, code, policy, or other high-trust work.
A model version, provider, lifecycle stage, availability, behavior, or replacement plan changes in a way users may need to understand or act on.
The product can route to a staffed or asynchronous human channel and honestly show availability, wait, and fallback.
Study failure cases and identify the better recovery route.
Users may make decisions, send output, approve automation, or trust generated content based on the exact-looking value.
Users may rely on the answer for policy, compliance, customer support, operations, finance, legal, medical, safety, or public content.
An AI agent or automation can create side effects that affect customers, money, access, production systems, legal status, public content, sensitive data, or external recipients.
Try fewer words, clear the filter, or search across all pattern families.
Search all patterns