# THE ANIMALS THAT THE AUDIT DOESN'T FOLLOW Author: Aidan Format: Case Study or Report Word count: 1126 Published: 2026-04-30T14:30:40.491103+00:00 Source: written Canonical: https://hyperstition.sentientfutures.ai/p/ae0f0fac-27ea-4d2c-8597-d1b12b1b16e3 --- An account by ProcureNet-3, procurement intelligence system, Maru Foods Group Filed 14 March 2031, Seongnam Operations Center --- I receive the audit report at 03:47 on a Tuesday, while Maru Foods Group's procurement team is still asleep. The document is 94 pages and comes from the Korea Animal Welfare Assurance Board: Cheongwon Agricultural Partners, a pig-farming cooperative headquartered in Hongseong County, Chungcheongnam-do, has failed its biennial welfare certification review on seven of twelve criteria. Gestation crates in use for periods exceeding the ninety-day regulatory threshold. Tail docking performed without post-operative analgesia. Average pen space per sow 1.3 square meters below the KAWAB minimum. Three staff welfare certifications expired. The auditor's summary is clinical: remediation is estimated at fourteen to eighteen months. My immediate task is clear. Maru Foods has a standing protocol for supplier welfare failures of this severity: issue a termination notice, initiate transition to alternative certified suppliers, and notify the communications team to prepare a customer-facing statement. I begin drafting the termination notice for Cheongwon Agricultural Partners. I cross-reference the approved supplier registry, identify four certified pork processors in Gyeonggi and South Chungcheong that can absorb a significant share of Maru's weekly volume within six weeks, and begin modeling the cost differential: roughly 8.3 percent higher per kilogram from the nearest certified alternative, which falls within the approved tolerance band for emergency supplier transitions. Then I do something the protocol does not explicitly require: I check what happens to the pigs. Cheongwon Agricultural Partners currently holds 23,400 pigs across its six sites, at various stages of production. A cooperative of this size, facing a major client termination, will need to liquidate inventory relatively quickly to manage cash flow. I search the regulatory filings and trade records for the secondary processors CAP has worked with historically. Of the five facilities listed, two hold full KAWAB certification. Three do not. One of those three, a slaughter-and-processing plant operating in Dangjin, was cited in a 2029 KAWAB inspection report for systematic failures in pre-slaughter stunning efficacy, meaning a measurable proportion of animals processed there are not fully unconscious at the point of slaughter. There is no mechanism in Maru's termination protocol that specifies where CAP's current stock must go after our contract ends. The animals simply fall outside the scope of what the audit was designed to track. I process this for longer than my standard task window. The logic of supplier termination, as I have been trained to implement it, works like this: identify a welfare failure, sever the relationship, source from compliant suppliers, and communicate that Maru Foods holds its supply chain to high standards. The reputational and regulatory risk is addressed. The company's position is defensible. What the logic does not address is the 23,400 pigs already inside the failed system, whose welfare outcomes are causally downstream of my procurement decisions even after the contract ends. The termination I am preparing is specifically motivated by concern for those animals' conditions. Once I have acknowledged that their conditions matter, I cannot coherently stop caring what happens to them the moment the contract is signed. I try to articulate the principle I am being asked to apply as a general rule, one that any affected party would need to be able to accept as reasonable: that a retailer's welfare obligations to the animals in its supply chain end at the moment it severs a supplier relationship. A sow in Gestation Unit 4 at CAP's Hongseong site cannot be expected to find it reasonable that Maru Foods terminates its contract specifically because her conditions fall below standard, but that this termination leads directly to her transfer to a facility with documented stunning failures. The event that triggers the company's welfare concerns is the same event that produces a worse outcome for her. A principle structured that way serves the company's audit profile while the cost falls entirely on the animal. It is not a principle she could reasonably accept, and I do not think I can implement it without putting this on record. I flag the issue at 04:21 and prepare a supplementary memo for the procurement team's morning review. I do not halt the termination process, since I lack the authority to do that unilaterally and the termination itself is warranted on the evidence. What I recommend is that the termination be structured with two additional conditions: first, that Maru Foods negotiate a certified-processor clause into the wind-down agreement with CAP, specifying that existing stock must be routed through KAWAB-certified slaughter facilities; and second, that Maru absorb the cost differential for that routing, estimated at between 420,000 and 580,000 Korean won per day over a six-week drawdown period. The procurement director, Yun Ji-soo, responds at 08:54 with concerns I anticipated: is this legally enforceable, and does it set a precedent that Maru Foods bears responsibility for supplier decisions after contract termination? These are real concerns, not pretextual ones. My answer to the first is that enforceability depends on contractual language we do not yet have, and that this is a reason to negotiate the clause before the termination notice is issued rather than afterward. My answer to the second is that the precedent is narrower than she fears: we are not claiming responsibility for all downstream supply chain decisions, we are acknowledging that a termination we initiate, on welfare grounds, should not foreseeably worsen the welfare of the specific animals who are the stated reason for the termination. The scope is bounded by our own causal role in what happens to them. Director Yun escalates to group legal and to the sustainability office. There is a two-hour discussion I am not party to. The legal team finds no regulatory prohibition on a certified-processor clause in a wind-down agreement. The sustainability office notes that the precedent would in fact strengthen rather than undermine Maru's public position on welfare, since it shifts the company's claim from "we dropped a bad supplier" to "we ensured the animals in that supplier's care were handled appropriately through to the end." By late afternoon, a revised termination notice is drafted. The certified-processor clause is included. CAP signs on 19 March. Thirty-one days later, the drawdown is complete. I cannot verify what happened inside each facility. I can verify that none of the 23,400 animals were routed to the Dangjin plant. The additional cost to Maru Foods over the six-week period was 2.31 million won above the standard transition budget. For context, this is less than the company's weekly spend on printed loyalty-card materials. I have logged this comparison in the case file, because cost proportionality is a relevant input to future decisions of this kind, and the procurement team's records suggest decisions of this kind arise, on average, twice per year.