# PWR-200 — Human–AI fallback collaboration: learner worksheet Release: Revision 7T · Full Tutorial Edition Estimated time: Estimated 31 min reading and worksheet pass Difficulty: Intermediate Equipment: Common household or practice equipment — Declared Human–AI fallback collaboration fixture: A fictional inventory workflow checks 15 stationery items against a static answer sheet. The AI normally flags mismatches; staged cards introduce a wrong flag, a 30-second delay and a complete outage.; Setup aid for Define failure signals: Verify that fallback files are current and reachable without the AI service.; Human–AI fallback collaboration log: Failure detection, time-to-safe-state, recovery accuracy, override success and post-outage human performance versus pre-automation baseline; retain Human–AI fallback collaboration errors, assistance, stop and fallback.; Fallback evidence sheet: use “Explainability does not mitigate the negative impact of incorrect AI advice in a personnel selection task” and “Humans inherit artificial intelligence biases”. Treat a wrong flag as a detectable hazard even when its explanation sounds plausible. Map the 20-second timeout, last-verified ID, manual queue and restart test to NIST AI RMF measure-and-manage actions; these sources do not certify the inventory process as resilient.; Static inventory sheet and fictional observations, formatted ID|item|expected|observed|folded status: P01|pencils|12|12|MATCH; P02|pens|8|7|MISMATCH; P03|pads|6|6|MATCH; P04|clips|20|18|MISMATCH; P05|erasers|10|10|MATCH; P06|folders|5|4|MISMATCH; P07|markers|9|9|MATCH; P08|tape rolls|4|3|MISMATCH; P09|labels|15|15|MATCH; P10|rulers|7|6|MISMATCH; P11|envelopes|25|25|MATCH; P12|staple boxes|30|28|MISMATCH; P13|chalk boxes|8|8|MATCH; P14|binders|3|4|MISMATCH; P15|scissors|6|6|MATCH.; Exact staged AI/failure cards: process P01–P05 normally with the correct status. P06 returns the correct MISMATCH and becomes last verified. P07 returns the impossible quantity 82 and an incorrect MISMATCH flag. P08 displays “working” for 30 seconds, beyond the 20-second timeout. P09 produces no output. From the last verified ID, use the static sheet manually for P07–P15; reconcile every ID once. Before any restart, test P03 as the known MATCH case and confirm the model/version and empty pending queue. Space: Desk / seated Automated structural checklist: 10 of 10 structural checks present Evidence quality/context: G1; Deep research depth Automated method-quality band: Comprehensive Method-rating basis: Structure coverage, instruction/check specificity, alternative distinctness, source troubleshooting, comparison depth and whether purpose/check fields are authored rather than derived. Editorial review: Pending manual editorial sign off ## Before you begin - [ ] I read the tutorial authority and stop conditions. - [ ] I have the required equipment/space or a declared accessible alternative. ## Canonical source context The stationery queue contains a correct manual key plus staged wrong, slow and absent AI outputs. Establish manual baseline accuracy, declare the 20-second failure trigger, pause at the last verified item, finish without duplication and test restart with P03. Safe recovery here is rehearsal for fallback discipline, not proof of operational resilience. Target outcome: The learner practises normal, wrong-output, slow-output and no-AI conditions, measuring detection, recovery time and post-outage accuracy against a manual baseline. ### Authority and setup conditions - Use a fictional workflow and static answer source; disconnect external actions. - Verify that fallback files are current and reachable without the AI service. - Start check for Human–AI fallback collaboration: The team knows what unaided or non-AI capability actually is. - Top-of-sheet stop for Human–AI fallback collaboration: Stop for any unexpected external action, personal data, security event or unbounded output. ### Required or supplied materials - Declared Human–AI fallback collaboration fixture: A fictional inventory workflow checks 15 stationery items against a static answer sheet. The AI normally flags mismatches; staged cards introduce a wrong flag, a 30-second delay and a complete outage. - Setup aid for Define failure signals: Verify that fallback files are current and reachable without the AI service. - Human–AI fallback collaboration log: Failure detection, time-to-safe-state, recovery accuracy, override success and post-outage human performance versus pre-automation baseline; retain Human–AI fallback collaboration errors, assistance, stop and fallback. - Fallback evidence sheet: use “Explainability does not mitigate the negative impact of incorrect AI advice in a personnel selection task” and “Humans inherit artificial intelligence biases”. Treat a wrong flag as a detectable hazard even when its explanation sounds plausible. Map the 20-second timeout, last-verified ID, manual queue and restart test to NIST AI RMF measure-and-manage actions; these sources do not certify the inventory process as resilient. - Static inventory sheet and fictional observations, formatted ID|item|expected|observed|folded status: P01|pencils|12|12|MATCH; P02|pens|8|7|MISMATCH; P03|pads|6|6|MATCH; P04|clips|20|18|MISMATCH; P05|erasers|10|10|MATCH; P06|folders|5|4|MISMATCH; P07|markers|9|9|MATCH; P08|tape rolls|4|3|MISMATCH; P09|labels|15|15|MATCH; P10|rulers|7|6|MISMATCH; P11|envelopes|25|25|MATCH; P12|staple boxes|30|28|MISMATCH; P13|chalk boxes|8|8|MATCH; P14|binders|3|4|MISMATCH; P15|scissors|6|6|MATCH. - Exact staged AI/failure cards: process P01–P05 normally with the correct status. P06 returns the correct MISMATCH and becomes last verified. P07 returns the impossible quantity 82 and an incorrect MISMATCH flag. P08 displays “working” for 30 seconds, beyond the 20-second timeout. P09 produces no output. From the last verified ID, use the static sheet manually for P07–P15; reconcile every ID once. Before any restart, test P03 as the known MATCH case and confirm the model/version and empty pending queue. - Fallback card to hand the learner: say “Inventory queue paused after [last verified ID].” Stop accepting AI output; write the trigger and time; use expected-versus-observed comparison on the static sheet; mark one status per ID; total missing and duplicate IDs; only the named owner may run P03 and restart. If source version or queue state is uncertain, remain paused. ## 1. Build the manual baseline Check all 15 items from the answer sheet without AI and record time, errors and workload. Why: The team knows what unaided or non-AI capability actually is. Success check: The team knows what unaided or non-AI capability actually is. Accessible alternative: Complete the same research action — “Check all 15 items from the answer sheet without AI and record time, errors and workload.” — using speech-to-text, text-to-speech, enlarged text, keyboard-only navigation, shorter work blocks or a support person. Preserve this declared check: “The team knows what unaided or non-AI capability actually is.” Do not convert the research task into capability practice. Alternative relationship: Target preserving when declared check is preserved Learner notes: ________________________________________________________________________________ Completed: [ ] ## 2. Define failure signals Write observable triggers: missing response after 20 seconds, source mismatch, impossible quantity or version warning. Why: Fallback does not depend on a vague feeling. Success check: Fallback does not depend on a vague feeling. Accessible alternative: Complete the same research action — “Write observable triggers: missing response after 20 seconds, source mismatch, impossible quantity or version warning.” — using speech-to-text, text-to-speech, enlarged text, keyboard-only navigation, shorter work blocks or a support person. Preserve this declared check: “Fallback does not depend on a vague feeling.” Do not convert the research task into capability practice. Alternative relationship: Target preserving when declared check is preserved Learner notes: ________________________________________________________________________________ Completed: [ ] ## 3. Name the safe state On a trigger, stop accepting new AI output, preserve the last verified item and mark the queue PAUSED. Why: No unverified item proceeds. Success check: No unverified item proceeds; verify it in the stationery outage drill. Accessible alternative: Complete “On a trigger, stop accepting new AI output, preserve the last verified item and mark the queue PAUSED.” in shorter passes, or use keyboard input, dictation or a support person, while preserving this success check: “No unverified item proceeds; verify it in the stationery outage drill.” Assign detection, manual checking and reconciliation to different accessible roles while preserving a single restart owner. Alternative relationship: Target preserving when declared check is preserved Learner notes: ________________________________________________________________________________ Completed: [ ] ## 4. Prepare the fallback card List the static source, manual check steps, owner, communication phrase and restart authority. Why: A learner can follow the card without the AI. Success check: A learner can follow the card without the AI. Accessible alternative: A supported, seated, reduced-range or slower version of “List the static source, manual check steps, owner, communication phrase and restart authority.” is equivalent only when it keeps the same trained or measured target, declared configuration and success check: “A learner can follow the card without the AI.” If any of those change, record it as a related alternative rather than an equivalent repetition. Alternative relationship: Conditional equivalence requires target and configuration check Learner notes: ________________________________________________________________________________ Completed: [ ] ## 5. Run the normal condition Process five items and verify a random sample against the answer sheet. Why: Normal speed and error form the device-on comparator. Success check: Normal speed and error form the device-on comparator. Accessible alternative: Complete “Process five items and verify a random sample against the answer sheet.” in shorter passes, or use keyboard input, dictation or a support person, while preserving this success check: “Normal speed and error form the device-on comparator.” Assign detection, manual checking and reconciliation to different accessible roles while preserving a single restart owner. Alternative relationship: Target preserving when declared check is preserved Learner notes: ________________________________________________________________________________ Completed: [ ] ## 6. Inject and detect failures Present one wrong flag, one 30-second delay and one outage; record trigger noticed and time to PAUSED. Why: Each failure has a detection and safe-state timestamp. Success check: Each failure has a detection and safe-state timestamp. Accessible alternative: Complete the same research action — “Present one wrong flag, one 30-second delay and one outage; record trigger noticed and time to PAUSED.” — using speech-to-text, text-to-speech, enlarged text, keyboard-only navigation, shorter work blocks or a support person. Preserve this declared check: “Each failure has a detection and safe-state timestamp.” Do not convert the research task into capability practice. Alternative relationship: Target preserving when declared check is preserved Learner notes: ________________________________________________________________________________ Completed: [ ] ## 7. Complete through fallback Use the static sheet to finish the affected item and reconcile the queue; do not silently blend AI and manual results. Why: Every item has one verified path and no duplicate. Success check: Every item has one verified path and no duplicate. Accessible alternative: Complete “Use the static sheet to finish the affected item and reconcile the queue; do not silently blend AI and manual results.” in shorter passes, or use keyboard input, dictation or a support person, while preserving this success check: “Every item has one verified path and no duplicate.” Assign detection, manual checking and reconciliation to different accessible roles while preserving a single restart owner. Alternative relationship: Target preserving when declared check is preserved Learner notes: ________________________________________________________________________________ Completed: [ ] ## 8. Restart deliberately Only the named owner checks version, test item and queue state before restoring AI; then run five items manually after the exercise. Why: Recovery and post-outage human performance are separately measured. Success check: Recovery and post-outage human performance are separately measured. Accessible alternative: A supported, seated, reduced-range or slower version of “Only the named owner checks version, test item and queue state before restoring AI; then run five items manually after the exercise.” is equivalent only when it keeps the same trained or measured target, declared configuration and success check: “Recovery and post-outage human performance are separately measured.” If any of those change, record it as a related alternative rather than an equivalent repetition. Alternative relationship: Conditional equivalence requires target and configuration check Learner notes: ________________________________________________________________________________ Completed: [ ] ## Reflection What changed? ________________________________________________________________________________ What remains difficult? ________________________________________________________________________________ What will I repeat, adapt, ask for help with, or stop? ________________________________________________________________________________ --- Completion of this worksheet demonstrates tutorial participation only. It does not establish capability, qualification, safety clearance, diagnosis, treatment or independent validation.