[Finance Fiction]Crack Protocol – Chapter 5: The Invisible Engine (chatgpt)
Theme: Finance Fiction (bank) + Suspense + Espionage + Workplace + The Dark Side of Human Nature + Anti-Money Laundering (AML)
⚠ Disclaimer
This is a work of fiction. While it draws inspiration from financial institutions, regulatory frameworks, and AML practices, it does not represent any real events or operations. The job titles, departmental structures, internal controls, review thresholds, and supervisory procedures have all been rewritten, merged, or fictionalized. This should not be taken as professional advice, investment guidance, or legal judgment. Any resemblance to real people, organizations, or systems is purely coincidental.
Theme: Finance Fiction (bank) + Suspense + Espionage + Workplace + The Dark Side of Human Nature + Anti-Money Laundering (AML)
Taipei City, Taiwan
Time: Friday 13:15 – Credit Department
Manager Zhang threw a stack of files onto the desk, his voice cold:
“Qilun, I’m assigning you these additional cases to review. Remember—don’t pull any more Sandbox tricks. I only want the tables exported from the production system. And your overtime hours? I’ll be monitoring those. Stay focused during work hours, and stop wasting time on irrelevant experiments.”
Colleagues stifled their laughter:
“Serves him right. He’ll spend the rest of his life buried in repetitive work.”
“Haha, with that sharp mouth of his, karma hit fast.”
Qilun simply replied with a quiet “Understood,” nothing more.
Time: Friday 13:25 – Desk Area
On his screen, Excel sheets flickered, dozens of workbooks running simultaneously.
All within IT’s access restrictions, but powered by his own VBA scripts, automating what could be automated.
Regulations and memos? He followed them to the letter.
Client: Company A (Textiles)
- Revenue growth: 25% (2024H1), industry average: 6–9%
- Gross margin consistently 10–12 percentage points above peers
- Staff count and utility usage barely changed—no matching increase in capacity
- Conclusion: Grey light (ambiguous observation zone)
- Suggested wording: Indicators above peers but still within acceptable range. Recommend supplemental major contracts to confirm reasonableness.
Client: Company B (Second-hand Market)
- Reported gross margin: 32%, industry average: 12–15%
- Accounts receivable turnover and cash flow alignment: normal
- Call notes lacked precision
- Conclusion: Grey light (ambiguous observation zone)
- Suggested wording: Recommend follow-up visit to clarify supply chain stability and substantiate margin sources.
Client: Company C (Construction)
- Financials show high leverage, DSCR near threshold
- Site visit notes: machinery loud, but few workers present
- Invoices all issued in the morning, contradicting “full-day operations” claim
- Conclusion: Red light (abnormal zone)
- Suggested wording: Calculation method inconsistent with credit manual. Please recalculate using standard definitions and attach details.
New Hire Rejection Cases
- Case 1 | Company A (Textiles)
- Staff report: DSCR = 1.46
- System recalculation: 1.18
- Cause: EBITDA substituted for CFO, principal repayment excluded
- Conclusion: Returned for resubmission
- Suggested wording: Calculation inconsistent with manual. Please recalculate per standard and attach details.
- Case 2 | Company B (Second-hand Market)
- Staff report: Gross margin = 88%
- System recalculation: 22%
- Cause: Quarterly figures multiplied ×4 to annualize
- Conclusion: Returned for resubmission
- Suggested wording: Quarterly data should not be annualized directly. Use rolling four quarters or actual annual figures, with supporting evidence.
- Case 3 | Company C (Construction)
- Staff report: Current ratio = 120%
- System recalculation: 85%
- Cause: Restricted deposits misclassified as current assets; unit mismatch (thousands vs. units)
- Conclusion: Returned for resubmission
- Suggested wording: Units or currency inconsistent. Please standardize and attach conversion details.
Overall Output Characteristics
- Green light: Normal cases (skipped entirely).
- Grey light: Language cautious, leaving room (“supplemental evidence required,” “recommend clarification”), avoiding false negatives.
- Red light: Evidence chain complete, pointed and irrefutable (as long as it matched the manual).
- Returned cases: Typical rookie mistakes (inconsistent methodology, improper annualization, unit errors), flagged concisely.
His program generated human-like ambiguity automatically.
Not “high risk → outright rejection.” But: “Needs supporting documentation to reduce misjudgment risk.”
This always left an escape hatch. Responsibility never stuck to him.
Built-in “Whitelist”
His script automatically filtered long-term and small-value clients—no red lights, only grey.
Outwardly, he called it “industry tolerance.” In truth, it was his own safety valve.
Auto-generated “Rejection Reason List”
His code produced standard fallbacks:
- “Calculation inconsistent with manual.”
- “Supplemental documentation required.”
- “Recommend follow-up visit.”
Each left room. No company ever received a direct “death sentence.”
Time: Friday 16:10 – Pantry
As he calmly brewed coffee, a colleague sneered:
“Don’t stay too late. I’m the one locking up.”
Qilun smiled faintly:
“Relax. I’ll be leaving on time today.”
That morning’s meeting still echoed in gossip.
A cluster of women poured coffee, whispering:
“Hey, what was all that math gibberish he spouted this morning?” “Who knows. Sounded like some nerdy role-play.” “But… when he spoke, his eyes were so focused. Kind of attractive, actually.” “Yeah, hard to talk to, but there’s depth in what he says.”
One burst out laughing:
“Attractive? Are you nuts? Imagine having his kid—wouldn’t it be slow-witted?” “He looks so prim. What if he’s secretly pervy, ready to pounce?”
The group roared with laughter.
Time: Friday 17:20 – Manager’s Desk
Manager Zhang skimmed through Qilun’s reports.
Many flagged issues were real, aligned with internal memos and the credit manual.
Eventually, Zhang barely dared to read them in detail—he simply set them aside.
Colleagues complained he wrote too much, “adding to everyone’s workload.” Some mocked his neatness, saying it was robotic.
Zhang’s stock response:
“As long as it’s compliant, no complaints matter.”
On the surface, respect for rules. But deep down he knew:
Credit is about performance. Meticulous review only trims losses.
Excuses, blame-shifting—that was daily bread.
Assigning more cases to Qilun was just a way to keep him “contained.”
Still, he wondered: This kid’s odd. If his family isn’t wealthy, why not just play it safe and climb the ladder slowly?
HR’s file said his background was ordinary.
Qilun’s Routine
By day, flawless performance.
Files checked clean, verification calls never missed.
To him, the procedures were too simple, too rigid—just routine.
What lit his eyes was only the numbers.
Whenever a formula or model demanded derivation, it was as if he was pulled into a black hole—obsessed to the point of stubbornness.
To others, he was a fanatic freak.
To himself, it was the only thing worth his focus.
Time: Friday 17:30 – Qilun’s Thoughts
He shut down the shared computer right at closing time.
A message blinked on his phone:
“Your article ‘Human-Like Credit Assessment’ reached 40,000 views.”
He knew the earnings were modest. But compared to office politics, this passive income was easy profit.
At the bank, reviews were just routine reports.
To him, they were experiments.
He was more focused and rigorous than anyone—yet he also knew better than anyone:
When the black swan arrived, all rules would collapse.
To him, the grey and red lights weren’t reports. They were illusions conjured by the banking system.
If you let things slide for the sake of conformity, the blame would come back to you.
He wasn’t covering for colleagues.
He was covering for the system.
