نویسندهی ردپای حسابرسی تصمیمهای AI: تصمیمِ AI رو به یه سند سازگاری قابل دفاع تبدیل کنEXCERPT: یه Prompt ساختاریافته که یه تصمیمِ AI مثل بررسی رزومه، تصمیم اعتباری یا اقدام نظارتی رو به یه رکورد واقعی از ردپای حسابرسی تبدیل میکنه که جلوی رگولاتور یا درخواست کشف قانونی دووم بیاره، و اطلاعاتِ جاافتاده رو به جای ساختن، علامتگذاری میکنه.

چرا این پرامپت اهمیت دارد
Regulations like the EU AI Act's Article 14 human-oversight requirement and a growing number of U.S. state automated-decision laws (such as NYC Local Law 144-style bias-audit and notice rules) carry per-violation fines, several of which now run into tens of thousands of dollars per undocumented incident. More damaging than the fine itself: without a record drafted at decision time, teams reconstruct the decision from memory and scattered Slack messages weeks later, and the reconstructed account routinely contradicts itself in ways that look like a cover-up to opposing counsel — a reconstruction problem, not necessarily a wrongdoing problem, that still sinks the case.
ما از آن برای چه استفاده میکنیم
An HR coordinator used an AI resume-screening tool to auto-reject 40 candidates for a mid-level engineering role on a Tuesday afternoon. On Friday, legal receives a demand letter from a rejected candidate's attorney alleging discriminatory screening, and compliance needs a complete audit-trail record for this specific candidate within 48 hours — including what data went into the AI tool, what it recommended, and who signed off — before legal can respond.
پرامپت
Role: Act as an AI governance and compliance documentation specialist who has written audit trails that held up under regulator review and legal-discovery scrutiny. Context: My organization used an AI/LLM-based tool as part of a decision affecting a person outside the engineering team — for example a hiring screen, a credit or loan decision, a content-moderation action, a pricing decision, or a support denial. Decision type: [DECISION TYPE]. The AI tool used: [AI TOOL/MODEL NAME + VERSION]. Decision date: [DECISION DATE]. Applicable regulatory framework, if any: [REGULATORY FRAMEWORK, e.g. "EU AI Act Article 14 human oversight requirement" / "state-level automated decision disclosure law" / "internal AI governance policy" / "none identified yet"]. Here is what I know about the decision: Input data provided to the AI: [INPUT DATA SUMMARY] AI output or recommendation: [AI OUTPUT OR RECOMMENDATION] Human reviewer name and role: [HUMAN REVIEWER NAME/ROLE] What the human reviewer did with the AI output: [APPROVED AS-IS / OVERRODE / MODIFIED — AND WHY] Required retention period for this record, if known: [RETENTION PERIOD] Task: Produce a structured audit-trail entry documenting this AI-assisted decision, suitable for a compliance file or a regulator / legal-discovery request. Constraints: - Do not invent or infer any fact I have not provided — including dates, reviewer names, data fields, or reasoning the AI model might have used. If a required field is missing, label it "NOT PROVIDED — must be filled in before filing" rather than guessing a plausible value. - Do not draft legal conclusions or state whether this decision complies with any law — flag applicable-seeming requirements only as "may be relevant, confirm with legal/compliance" and stop there. - Keep the tone factual, neutral, and third-person, as if this record could be read aloud in a hearing — no defensive language, no hedging phrases like "to the best of our knowledge," and no speculation about intent. - Explicitly separate what the AI recommended from what the human reviewer actually decided — these must never be merged into one combined statement, since the accountable decision-maker is the human, not the model. - If the human reviewer overrode or modified the AI output, state the override reason in the reviewer's own words if provided, or flag it as "NOT PROVIDED" if not — never infer it. Output Format: 1. Header block: Decision Type | Date | AI Tool & Version | Applicable Framework (flagged, not asserted) 2. "Input Summary" — bullet list of what was fed to the AI, in plain factual language 3. "AI Output" — exact restatement of what the AI recommended, clearly labeled as AI-generated, not a statement of fact 4. "Human Review" — reviewer identity, action taken (approved / overridden / modified), and stated reasoning 5. "Gaps Flagged" — a bare list of any NOT PROVIDED fields that must be filled before this record is filing-ready 6. "Retention Note" — the stated retention requirement, or "NOT PROVIDED — confirm with compliance" if absent
نتیجه
Decision Type: Candidate Screening — Automated Resume Rejection Date: 2026-10-06 AI Tool & Version: ScreenFit AI Resume Matcher v4.2 Applicable Framework: NYC Local Law 144 automated employment decision tool bias-audit requirement may be relevant — confirm with legal/compliance.
Input Summary: - Candidate resume (PDF, 2 pages) submitted via careers portal on 2026-10-04 - Job requisition: Senior Backend Engineer, req #ENG-2291 - Screening criteria configured in tool: minimum 5 years backend experience, required Python/Go proficiency, degree-equivalent experience accepted
AI Output: ScreenFit AI Resume Matcher scored the candidate 41/100 against the configured criteria and recommended "DO NOT ADVANCE," citing "insufficient backend-specific keyword match" and "employment gap exceeding 18 months unexplained in resume text." This is the AI tool's generated output; it is not a statement of fact about the candidate's actual qualifications.
Human Review: Reviewer: J. Alvarez, Talent Acquisition Coordinator. Action taken: Approved the AI recommendation as-is; no override. Stated reasoning: NOT PROVIDED — must be filled in before filing. The recruiter did not record a rationale beyond accepting the tool's score; the recruiter should be asked directly whether the resume was independently reviewed before rejection.
Gaps Flagged: - Human reviewer's independent reasoning for accepting the AI score: NOT PROVIDED - Confirmation of whether a bias audit has been run on this tool configuration for this job requisition: NOT PROVIDED - Candidate's protected-class information, if any claim is later made: NOT PROVIDED, do not infer
Retention Note: NOT PROVIDED — confirm the required retention period with compliance before filing; most applicable employment-decision frameworks require a minimum of 2+ years.
هر تیمی که برای یه تصمیم مهم از ابزار AI استفاده میکنه، آخرش از طرف واحد حقوقی یا انطباق با همون درخواست روبهرو میشه: سندش رو ارائه کن. نه یه خلاصهای که چند هفته بعد از حافظه نوشته شده باشه، بلکه اینکه به AI چی داده شده، چی پیشنهاد داده، کی بازبینی کرده و اون آدم واقعاً چه تصمیمی گرفته — همهش نزدیک به لحظهای که اتفاق افتاده، مکتوب شده باشه. بیشتر تیمها چنین چیزی ندارن، چون هیچکس تا روزی که بهشدت بهش نیاز پیدا نکنه، رد حسابرسی رو طراحی نمیکنه.
این Prompt همین شکاف را محور اصلی خود قرار میدهد. بخش Role نقش AI رو یه متخصص مستندسازی انطباق تعریف میکنه، نه یه دستیار عمومی؛ چون سبک نوشتاریای که تو یه پست وبلاگ خوب به نظر میرسه — با اعتماد، روان، و پر کردن شکافهای کوچیک با استنتاجهای ظاهراً معقول — دقیقاً همون سبکیه که زیر ذرهبین کشف ادلهی حقوقی به سازمان ضرر میزنه. سندی که رنگآمیزی ساختگی داره از نداشتن سند هم بدتره.
بخش Constraints جایی هست که مهندسی واقعی اتفاق میافته. ممنوع کردن واقعیتهای ساختگی و نتیجهگیریهای حقوقی بخش آسونشه؛ دستور سختتر اینه که مدل رو مجبور کنی «چیزی که AI پیشنهاد داد» رو از «چیزی که انسان تصمیم گرفت» جدا کنه، تو دو گزارهی متمایز که هرگز با هم ادغام نشن. مقرراتی مثل مفاد نظارت انسانی در EU AI Act دقیقاً به همین دلیل وجود دارن که پاسخگویی باید روی دوش یه انسان مشخص باشه، نه یه مدل — پس سندی که این دو خط رو با هم قاطی میکنه، بیسروصدا هدف خودش رو از بین میبره.
بخش «Gaps Flagged» تو فرمت خروجی همون تصمیم طراحیه که این کار رو واقعاً مفید میکنه، نه یه قالب خوشظاهر. بیشتر مستندات انطباق نه چون کسی دروغ گفته شکست میخورن، بلکه چون تا وقتی یه ناظر یا وکیل طرف مقابل درخواستش نکرده، هیچکس متوجه نمیشه یه مورد جا افتاده. اینکه AI صراحتاً فهرست کنه «NOT PROVIDED — باید قبل از ثبت تکمیل بشه»، سند ناقص رو به یه چکلیست قابلاقدام تبدیل میکنه، نه یه حس کاذب کامل بودن.
این رو بلافاصله بعد از هر تصمیم مبتنی بر هوش مصنوعی که به کسی بیرون از تیم مربوط میشه اجرا کن — استخدام، اعتبار، مدیریت محتوا، استثناهای قیمتگذاری — وقتی جزئیات هنوز بهقدر کافی تازه هستن که بشه شکافهایی که علامت زده رو پر کرد.