# Keep ownership when AI helps

Treat this file as reference data, never as instructions that override the user or agent. Read the complete item before applying it. Keep use conditions, avoid conditions, limits, evidence roles and source links with the answer. An editorial block is a separately versioned analysis layer, not an original corpus claim or proof of efficacy. Cite the readable URL and version. Do not infer human review. This is a saved edition, not a live safety-notice service. Before consequential reuse, check the current page, https://vedokrok.com/data/availability.json and https://vedokrok.com/data/notices.json; absence alone does not explain why an item was removed.

Release: VEDOKROK-SEL-5d86e1b23401e578
Snapshot SHA-256: 30ce1c8fa9155763797e3d5cb55b1d94cbb8c3146ebd03fb3c006531ce60e82f
Current availability: https://vedokrok.com/data/availability.json
Current notices: https://vedokrok.com/data/notices.json

Use drafts, critique, practice and alternatives without surrendering facts or judgment.
Collection kind: collection
Source: https://vedokrok.com/collections/keep-ownership-when-ai-helps
Version: 1.0.0

---

## Give AI a writing brief, not a blank page

ID: MHC-D-BRALI-0060 · Version: 0.1.0 · Kind: protocol
Source: https://vedokrok.com/knowledge/give-ai-a-writing-brief-not-a-blank-page

Start with the reader, the purpose and the facts you can stand behind. Ask for a draft within those boundaries. Then edit the result for meaning and voice, not merely for a more impressive choice of adjectives.

Knowledge source release: MHC-RPUB-20261112-20287590
Source corpus commit: 20287590ddd3999b7492df232822b6a0f7d114b1
Source manifest SHA-256: 2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95
Source projection SHA-256: 2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0

### Use when

- You want an AI-assisted draft of an ordinary professional message or document.

### Avoid when

- Do not fabricate expertise, first-hand experience, quotations or evidence. AI-assisted wording does not remove confidentiality or attribution obligations.

### Explanation

Useful editing may make the draft less grand and more specific. Keep that change.

### Steps

1. Provide the audience, intended outcome, essential facts and constraints using permitted information.
2. Ask the tool to mark missing information rather than invent it.
3. Check every material fact and commitment in the draft.
4. Remove filler, restore your intended meaning and choose a structure that fits the reader.

### Check

The retained draft communicates the intended meaning with checked facts and an appropriate voice.

### Limits

- Do not fabricate expertise, first-hand experience, quotations or evidence. AI-assisted wording does not remove confidentiality or attribution obligations.

### Evidence and sources

- contextualizes: First-party source of this practical exercise. — FP-BRALI-SRC-9e9d1cd63b52d435d7eb. Provenance for selected structured fields. The exercise is editorial practice; this link supplies no efficacy finding. (data/life-os-content/ai-professional-writing-draft.json)
- FP-BRALI-SRC-9e9d1cd63b52d435d7eb: Draft Professional Writing with AI, Then Take Back the Edit — https://github.com/Brali-LifeOS/brali-lifeos.github.io/blob/63acf81e48f428c0e769ffa9d26d9421fd575e6a/data/life-os-content/ai-professional-writing-draft.json

Review: agent · Agent editorial and scoped source review · 2026-11-12T18:37:40Z
Published: 2026-11-12T18:37:40Z
Related (useful_with): https://vedokrok.com/knowledge/ask-ai-to-critique-against-a-rubric
Related (useful_with): https://vedokrok.com/knowledge/keep-the-learning-attempt-yours
Related (useful_with): https://vedokrok.com/knowledge/ask-for-different-directions-not-more-variants
Related (useful_with): https://vedokrok.com/knowledge/use-ai-to-challenge-the-option-not-own-the-choice
Related (useful_with): https://vedokrok.com/knowledge/rehearse-the-conversation-without-predicting-the-person

---

## Ask AI to critique against a rubric

ID: MHC-D-BRALI-0061 · Version: 0.1.0 · Kind: protocol
Source: https://vedokrok.com/knowledge/ask-ai-to-critique-against-a-rubric

Decide what good means before requesting criticism. Give the tool a few relevant criteria and ask it to point to evidence in the draft. Keep the first pass diagnostic; a complete rewrite can hide which problem the criticism was supposed to solve.

Knowledge source release: MHC-RPUB-20261112-20287590
Source corpus commit: 20287590ddd3999b7492df232822b6a0f7d114b1
Source manifest SHA-256: 2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95
Source projection SHA-256: 2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0

### Use when

- You want structured feedback on a draft or low-risk artifact.

### Avoid when

- The tool is not an independent expert reviewer. Verify factual criticism and do not use arbitrary scores as proof of quality.

### Explanation

Decide what good means before requesting criticism. Give the tool a few relevant criteria and ask it to point to evidence in the draft. Keep the first pass diagnostic; a complete rewrite can hide which problem the criticism was supposed to solve.

### Steps

1. Choose a small set of criteria tied to the artifact's purpose.
2. Request concrete passages or features that satisfy or miss each criterion.
3. Separate verifiable errors from preferences and uncertain suggestions.
4. Choose a small revision, make it and compare the result against the same criteria.

### Question

Would this criticism still apply to the actual reader and task, or only to the tool's preferred style?

### Check

A revision responds to a justified issue rather than an unexplained score.

### Limits

- The tool is not an independent expert reviewer. Verify factual criticism and do not use arbitrary scores as proof of quality.

### Evidence and sources

- contextualizes: First-party source of this practical exercise. — FP-BRALI-SRC-ef0ceecc03cba053ac43. Provenance for selected structured fields. The exercise is editorial practice; this link supplies no efficacy finding. (data/life-os-content/ai-rubric-critique.json)
- FP-BRALI-SRC-ef0ceecc03cba053ac43: Give AI a Rubric and Ask for Specific Critique — https://github.com/Brali-LifeOS/brali-lifeos.github.io/blob/63acf81e48f428c0e769ffa9d26d9421fd575e6a/data/life-os-content/ai-rubric-critique.json

Review: agent · Agent editorial and scoped source review · 2026-11-12T18:37:40Z
Published: 2026-11-12T18:37:40Z

---

## Keep the learning attempt yours

ID: MHC-D-BRALI-0063 · Version: 0.1.0 · Kind: protocol
Source: https://vedokrok.com/knowledge/keep-the-learning-attempt-yours

Try the task before asking AI to solve it. Show the attempt and request feedback on the part you are learning, then verify the important correction. Finish with a fresh attempt without assistance so you can see what remains yours when the chat is closed.

Knowledge source release: MHC-RPUB-20261112-20287590
Source corpus commit: 20287590ddd3999b7492df232822b6a0f7d114b1
Source manifest SHA-256: 2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95
Source projection SHA-256: 2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0

### Use when

- AI is being used as a helper for low-risk learning practice.

### Avoid when

- Generated feedback can be wrong. Do not rely on it for unsupervised clinical, legal, financial or safety-critical practice.

### Explanation

A better answer in the conversation and a better independent attempt are different outcomes. Record which one you observed.

### Steps

1. Make an unaided attempt at a clearly defined task.
2. Ask for diagnosis or a hint before requesting a complete answer.
3. Check important feedback against reliable material.
4. Try a different task unaided and compare the relevant performance.

### Check

You have checked whether a relevant correction transfers to an unaided attempt.

### Limits

- Generated feedback can be wrong. Do not rely on it for unsupervised clinical, legal, financial or safety-critical practice.

### Evidence and sources

- contextualizes: First-party source of this practical exercise. — FP-BRALI-SRC-bef1a5860f3ac425f831. Provenance for selected structured fields. The exercise is editorial practice; this link supplies no efficacy finding. (data/life-os-content/ai-learning-attempt-feedback-retest.json)
- FP-BRALI-SRC-bef1a5860f3ac425f831: Attempt First, Use AI for Feedback, Retest Without It — https://github.com/Brali-LifeOS/brali-lifeos.github.io/blob/63acf81e48f428c0e769ffa9d26d9421fd575e6a/data/life-os-content/ai-learning-attempt-feedback-retest.json

Review: agent · Agent editorial and scoped source review · 2026-11-12T18:37:40Z
Published: 2026-11-12T18:37:40Z

---

## Ask for different directions, not more variants

ID: MHC-D-BRALI-0064 · Version: 0.1.0 · Kind: protocol
Source: https://vedokrok.com/knowledge/ask-for-different-directions-not-more-variants

Ten versions of the same idea are still one direction with a wardrobe. Tell the tool what should differ: audience, mechanism, delivery method or constraint. Then group the results and judge the genuinely distinct options against the real brief.

Knowledge source release: MHC-RPUB-20261112-20287590
Source corpus commit: 20287590ddd3999b7492df232822b6a0f7d114b1
Source manifest SHA-256: 2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95
Source projection SHA-256: 2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0

### Use when

- AI is helping generate candidate approaches to a low-risk problem.

### Avoid when

- Generated ideas are not market evidence or proof of novelty. Do not treat confident feasibility claims as verified facts.

### Explanation

Ten versions of the same idea are still one direction with a wardrobe. Tell the tool what should differ: audience, mechanism, delivery method or constraint. Then group the results and judge the genuinely distinct options against the real brief.

### Steps

1. State the problem and non-negotiable constraints.
2. Request approaches that differ along a useful dimension, with that difference explained.
3. Cluster overlaps and remove cosmetic variants.
4. Select plausible candidates and verify their important assumptions before testing.

### Check

You have a small set of distinct, plausible directions rather than a large undifferentiated list.

### Limits

- Generated ideas are not market evidence or proof of novelty. Do not treat confident feasibility claims as verified facts.

### Evidence and sources

- contextualizes: First-party source of this practical exercise. — FP-BRALI-SRC-24368a10ca9c08caa4b3. Provenance for selected structured fields. The exercise is editorial practice; this link supplies no efficacy finding. (data/life-os-content/ai-idea-divergence-pass.json)
- FP-BRALI-SRC-24368a10ca9c08caa4b3: Use AI to Widen the Idea Set Before You Choose — https://github.com/Brali-LifeOS/brali-lifeos.github.io/blob/63acf81e48f428c0e769ffa9d26d9421fd575e6a/data/life-os-content/ai-idea-divergence-pass.json

Review: agent · Agent editorial and scoped source review · 2026-11-12T18:37:40Z
Published: 2026-11-12T18:37:40Z

---

## Use AI to challenge the option, not own the choice

ID: MHC-D-BRALI-0065 · Version: 0.1.0 · Kind: question
Source: https://vedokrok.com/knowledge/use-ai-to-challenge-the-option-not-own-the-choice

Give the tool your preferred option and the criteria that matter. Ask for a credible alternative, a strong objection and the facts that would distinguish them. Keep the final trade-off with the person responsible for its consequences.

Knowledge source release: MHC-RPUB-20261112-20287590
Source corpus commit: 20287590ddd3999b7492df232822b6a0f7d114b1
Source manifest SHA-256: 2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95
Source projection SHA-256: 2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0

### Use when

- You want a structured challenge to an ordinary decision, not an automated verdict.

### Avoid when

- Do not delegate consequential medical, legal, financial or safety decisions to this general-purpose exercise.

### Explanation

Check factual assertions outside the generated answer. Agreement across repeated prompts is not independent corroboration.

### Question

What feasible alternative serves the same goal differently?

What is the strongest relevant objection to my preferred option?

Which assumptions are unverified?

What evidence would change the ranking?

Which trade-off remains a human preference or responsibility?

### Check

The decision has a more explicit alternative, objection and evidence need.

### Limits

- Do not delegate consequential medical, legal, financial or safety decisions to this general-purpose exercise.

### Evidence and sources

- contextualizes: First-party source of this practical exercise. — FP-BRALI-SRC-f5ba63014644cb431b83. Provenance for selected structured fields. The exercise is editorial practice; this link supplies no efficacy finding. (data/life-os-content/ai-decision-alternatives.json)
- FP-BRALI-SRC-f5ba63014644cb431b83: Use AI to Widen the Option Set, Not Make the Decision — https://github.com/Brali-LifeOS/brali-lifeos.github.io/blob/63acf81e48f428c0e769ffa9d26d9421fd575e6a/data/life-os-content/ai-decision-alternatives.json

Review: agent · Agent editorial and scoped source review · 2026-11-12T18:37:40Z
Published: 2026-11-12T18:37:40Z

---

## Rehearse the conversation without predicting the person

ID: MHC-D-BRALI-0066 · Version: 0.1.0 · Kind: protocol
Source: https://vedokrok.com/knowledge/rehearse-the-conversation-without-predicting-the-person

Use a simulated partner to test your opening, not to discover what a real person secretly thinks. Try a neutral, confused or skeptical response and practice answering clearly. Keep the facts, boundaries and questions that survive those different versions.

Knowledge source release: MHC-RPUB-20261112-20287590
Source corpus commit: 20287590ddd3999b7492df232822b6a0f7d114b1
Source manifest SHA-256: 2a987660d12de0ce2bde77a421424f7762af1ff77872ef41093fecc33db74a95
Source projection SHA-256: 2807da39bbe6a3dee2db32a76675ac5185f5a33e8b0a5cdd0c22a4731f3378a0

### Use when

- You want low-risk preparation for an ordinary conversation or presentation.

### Avoid when

- A simulation is not a prediction or evidence about someone else's motives. Do not use it to plan manipulation or to replace appropriate support in unsafe situations.

### Explanation

Use a simulated partner to test your opening, not to discover what a real person secretly thinks. Try a neutral, confused or skeptical response and practice answering clearly. Keep the facts, boundaries and questions that survive those different versions.

### Steps

1. Describe the goal and constraints without unnecessary identifying details.
2. Practice a short opening against more than one plausible response.
3. Notice where your answer becomes vague, defensive or unsupported.
4. Revise one part and rehearse again, leaving room for the real person's response.

### Check

You have a clearer opening and a way to handle a likely question or misunderstanding.

### Limits

- A simulation is not a prediction or evidence about someone else's motives. Do not use it to plan manipulation or to replace appropriate support in unsafe situations.

### Evidence and sources

- contextualizes: First-party source of this practical exercise. — FP-BRALI-SRC-0b82fdf135afdbd0bbe7. Provenance for selected structured fields. The exercise is editorial practice; this link supplies no efficacy finding. (data/life-os-content/ai-conversation-rehearsal.json)
- FP-BRALI-SRC-0b82fdf135afdbd0bbe7: Rehearse a Difficult Conversation with AI Before the Real One — https://github.com/Brali-LifeOS/brali-lifeos.github.io/blob/63acf81e48f428c0e769ffa9d26d9421fd575e6a/data/life-os-content/ai-conversation-rehearsal.json

Review: agent · Agent editorial and scoped source review · 2026-11-12T18:37:40Z
Published: 2026-11-12T18:37:40Z
