Listen Get

The Slaughterhouse Was a Workaround

Keywords

war-on-disease, 1-percent-treaty, medical-research, public-health, peace-dividend, decentralized-trials, dfda, dih, victory-bonds, health-economics, cost-benefit-analysis, clinical-trials, drug-development, regulatory-reform, military-spending, peace-economics, decentralized-governance, wishocracy, blockchain-governance, impact-investing

You wanted a molecule. Protein, mostly. Some fat, a little iron, a dab of B12. So you built a machine to get it.

The machine is a cow. It takes two years to compile, runs on a field of soybeans, returns about 3% of the calories you feed it10, occupies more land than any other structure your species operates, emits a greenhouse gas as a standard feature, occasionally ships a new pandemic, and screams at the end. Then you extract the molecule, which was available in the soybeans the entire time.

On my planet, we call this a workaround. You could not yet synthesize food proteins directly, so you domesticated a self-replicating bioreactor that came pre-installed in a nervous system. It was a clever hack for the Neolithic. You have since sequenced the genome, folded the proteins, and won a Nobel Prize for building new ones from scratch182. The workaround is now the only part of your civilization still running on the original hardware, and the original hardware can feel pain.

The Middleware

Every engineer eventually finds the layer that does nothing but add cost. In your food system, that layer is the animal.

Run the numbers as a systems problem. Livestock occupies 80% of all farmland93, drinks its way through most of the 70% of your fresh water that agriculture withdraws2, and in return delivers 18% of your calories and 37% of your protein94. You have assigned the majority of your planet’s usable surface to the least productive tenant on it. If a data center used 80% of your power grid to run 18% of your compute, you would fire the person who built it, and then you would fire the person who defended it.

The inefficiency is not an accounting quirk. It is thermodynamics. You grow a plant, which already made the protein. Then you feed the plant to an animal, which spends most of those calories being a warm, mobile, frightened animal, and stores a fraction as meat. Getting protein from beef instead of peas emits roughly ninety times more carbon per gram183. You are not producing food. You are laundering it through a second organism and paying a ninety-to-one spread for the privilege. The animal is the middleman, and like all middlemen who add no value, it survives only because everyone upstream is subsidized and nobody downstream can see the invoice.

The Bill Nobody Prints

Here is the invoice.

Your species kills 83 billion animals on land every year for food46. Chickens dominate the count, at roughly 140,000 per minute, which is a number that keeps happening while you read this sentence, and the next one, and every sentence you will ever read184. In the water it is worse by an order of magnitude: on the order of 1.1 trillion wild-caught fish a year, a range so large because nobody bothered to count creatures they weigh by the ton6, plus another hundred billion farmed185. In the United States, 99% of these animals live their entire lives inside a factory farm45, a phrase your marketing departments worked hard to keep off the packaging.

By raw headcount of feeling creatures, this is the largest thing your species does. It is larger than war, larger than disease, larger than every other source of suffering combined, and it is the one your ledgers refuse to book. Your economists price the meat. They never price the animal’s Tuesday.

I want to be precise about what this manual can and cannot promise you here. The machine described in the rest of this book optimizes two numbers, both of them human: median healthy lifespan and median income. Animal suffering is not in that objective function. It is a third axis, and adding it is a real argument your species has not finished having. So the case that follows does not depend on you weighting a chicken’s life at anything above zero. Every other benefit, the land, the water, the pandemics, the money, the human deaths, lands whether or not you care about the chicken. The chicken is simply the largest item on an invoice that also happens to be itemized in your own currency.

The Pandemic Startup

If you did want a business that manufactured pandemics, you would build the current one.

Take tens of billions of genetically similar animals, immune systems identical enough to share a vulnerability, pack them at the densities a factory farm requires, keep them alive on a constant drip of antibiotics, and place the whole operation next to humans. Roughly 83 billion birds a year is a lot of serial passage for a virus to practice on. Three out of four new human diseases already come from animals186; you have built the ideal incubator and then acted surprised each time it ships.

The antibiotics are their own slow catastrophe. About 80% of all antibiotics sold in the United States go to animal agriculture, and about 70% of those are medically important to humans187. You are spending the effectiveness of your miracle drugs to make cheap chicken slightly cheaper. Antimicrobial resistance already causes about 1.27 million deaths (95% CI: 911 thousand deaths-1.71 million deaths) a year directly3; one government-commissioned projection, contested and drawn from the pessimistic end of its own range, put the toll at ten million a year and one hundred trillion dollars by 2050 if nothing changes188. Treat that as a scenario, not a forecast. Treat the incubator as a fact.

You have a recent price tag for one pandemic that got out. Two economists put the cost of COVID-19 to the United States alone at more than sixteen trillion dollars, about ninety percent of a year’s entire economy189. The food system did not cause that particular one, but it is the sector most likely to cause the next, and you are paying it a subsidy to keep trying.

Three Ways to Skip the Animal

The good news is the part your species keeps burying under hype. You do not need the animal to get the molecule, and this is not speculative. You already do it, at commodity scale, for the parts of the animal you found convenient to replace first.

Brew it (precision fermentation). You take the gene for the protein you want, put it in a microbe, and let the microbe make the protein in a tank, exactly as you already make insulin instead of harvesting it from pig pancreases. About 90% of the cheese in the United States is already set with rennet brewed this way rather than scraped from a calf’s stomach, and has been since 1990, and nobody marched190. The enzyme is molecularly identical; the calf is simply absent. Companies now brew the actual whey protein from milk with no cow attached. This is the mature technology. It works for enzymes and specific proteins today; the open problem is doing it cheaply enough for bulk protein, not whether it works.

Grow it (cultivated meat). You take animal cells and grow the muscle without the animal around it. Here I will do something the pitch decks will not: tell you it is contested. The optimistic techno-economic analysis, commissioned by an advocacy group, projects costs near five to six dollars a kilogram by 2030 in its best case191. A skeptical peer-reviewed analysis concluded the economics “would likely preclude the affordability of their products as food,” pricing cell mass around thirty-seven dollars a kilogram and doubting it falls far enough192. Both cannot be right. The disagreement turns on the cost of growth media and the performance of very large bioreactors, which are engineering questions, not miracles. A citable report should say which number is a hope and which is a measurement. This one is a hope with a serious objection attached.

Design it (plants and computation). The oldest route is the plainest: get the protein straight from the plant that made it, and use protein engineering to make it behave like the animal version. The tools arrived in 2024 with a Nobel Prize, for predicting how proteins fold and for building entirely new ones to specification182. Most food work today still discovers useful natural proteins rather than designing novel ones, so do not oversell the designer-molecule story yet; the discovery half alone is already enough to replace a great deal of what the animal was for.

One honest caveat that protects everyone building here: the current alternative-protein market is in a downturn, not a boom. Investment fell 27% in 2024, and plant-based venture funding fell 64%193. The forecasts promising the cow’s extinction by 2030 came from a think tank whose 2030 predictions are already testable and already wrong194. The platform is real. The timelines are marketing. Build on the platform; discount the timelines.

The Land Comes Back

Now the number that should reorganize your priorities.

If your species stopped eating animals, the farmland you would need shrinks by more than 75%, an area the size of the United States, China, the European Union, and Australia combined, and you would still feed everyone47. That land is not abstract. Cattle pasture alone drives 41% of tropical deforestation195; animal-based foods account for 57% of the emissions from growing food at all196, and growing food is itself about 26% of every greenhouse gas your species emits50. Delete the middleware and you are not just saving money. You are handing back a third of the habitable surface of the planet, which promptly reforests itself and starts pulling your carbon out of the sky for free, as it did for the several billion years before you fenced it.

This is the rarest kind of fix: one intervention that reduces land use, water use, emissions, deforestation, pandemic risk, antibiotic collapse, diet-driven disease (poor diet is tied to about 11 million deaths (95% CI: 10 million deaths-12 million deaths) a year34), and the largest reservoir of suffering on Earth, all at once, by removing a component rather than adding one. Your species usually has to trade one problem for another. Here the trade is: give up the workaround, keep the food.

Why the Machine Builds This Anyway

You do not need a new government agency for this, and you should distrust anyone who tries to sell you one. The optimized food system is not a fourteenth item in the Government Replacement Suite. It is an output of the machinery already in this book.

Point the Optimal Budget Generator197 at agriculture and the first thing it flags is the money you already spend keeping the old machine upright. In the United States alone, agricultural subsidies run a deadweight loss around $75 billion (95% CI: $50 billion-$120 billion) a year155, much of it propping up exactly the animal-feed economics that make the workaround look cheap. The whole global food system carries roughly $12 trillion a year in hidden costs, sitting off your books, paid by your lungs, your hospitals, and your grandchildren51. The Optimal Budget Generator does not moralize about any of this. It notices that a dollar moved from subsidizing the least efficient calories on Earth to funding the dFDA198,199-style trials and prizes that make the new production lines cheap returns more health and more wealth than the dollar was doing before. Then it hands that reallocation to a lobbyist, and the math does the selling.

That is the mechanism. The subsidy stops flowing to the middleware. The redirected money makes fermentation and cultivation and plant protein cheaper faster. The moment the molecule is cheaper without the animal than with it, the market does the rest, the way it always does once the price is finally telling the truth. The optimized civilization does not ban the cow. It stops paying you to keep one.

Reviews from the Competition

“One star. They kept the protein and cancelled me. I was the protein’s ride. I had a whole personality.” (The Cow)

“We had a beautiful thing going: dense flocks, shared immune systems, humans next door, antibiotics on tap. A perfect launchpad. Then they moved production into a sealed tank with no lungs in it. Devastating.” (Influenza)

“They reforested my pasture. Now it’s just trees, breathing, pulling carbon, doing nothing for the shareholders. Nature has no business model.” (The Quarterly Report)

Build Sheet

Loop role: replace the least-optimized production system on the planet. You are not lobbying for this one; you are building it, and the machine’s job is to make your unit economics win.

  • What you are building: One animal-free production line that ships a molecule people already buy. Pick the route that matches your capital and your chemistry: a fermentation tank brewing a specific protein or enzyme (mature, fundable now), a cultivation process (higher risk, honest about it), or a plant-plus-engineering product (lowest tech risk, hardest on taste). Ship the ingredient first; the branded food comes after the molecule is validated and cheap.
  • Parts required: A target protein with a known sequence. A production host (microbe, cell line, or crop). Access to protein-folding and expression-prediction tools, now largely public since the 2024 prize182. And the demand signal the Optimal Budget Generator creates when it moves the $75 billion (95% CI: $50 billion-$120 billion) subsidy off your competitor and toward the trials that de-risk you.
  • Specifications: Molecular identity to the animal version where the function requires it (an enzyme, a whey, a heme). Honest lifecycle accounting: your land and water advantage is real and large; your energy footprint is only an advantage if your power is clean, so buy clean power and put the number in your own deck before a skeptic puts it in theirs.
  • Testing your installation: (1) Cost per kilogram of finished protein, printed next to the animal incumbent, updated quarterly, no asterisk. (2) A blind sensory panel your product survives. (3) A published lifecycle assessment you did not commission from your own marketing department.
  • Parts cost: Fermentation of a specific protein is fundable at startup scale today; the enzyme precedent proves the platform. Cultivation needs patient capital and a media-cost breakthrough you should name explicitly, not assume. Do not quote a 2030 cost curve as if it were a purchase order.
  • First bolt (no permission required): Brew or grow one gram of one protein that is molecularly identical to an animal product, and measure what it cost. Publishing that single honest number, next to the incumbent, moves more capital than any forecast, because the entire sector’s credibility problem is that it keeps quoting the hope instead of the measurement. Be the one who quotes the measurement.
  • Troubleshooting:
Symptom Fix
“Lab-grown food is unnatural.” So is insulin from a tank, rennet from a microbe, and 90% of your cheese, all of which you already eat without incident190. “Natural” is the animal that gives you the pandemic; “unnatural” is the version that does not. Pick your unnatural.
“Cultivated meat can’t hit cost parity.” Maybe not soon, and the skeptic has a peer-reviewed case192. So do not lead with cultivated meat. Lead with fermentation, which already hit parity for enzymes, and let cultivation mature behind it. The platform does not stand or fall on the hardest route.
“This destroys farmers and rural economies.” The subsidy destroys farmers; it pays the largest operations to overproduce the least valuable calories. Redirect the same money to feedstock crops, fermentation facilities, and land stewardship, and the acreage that stops growing animal feed grows either food people eat or forest that pays a carbon yield. The transition is a line item, and the machine funds it deliberately, not by accident.
“The alt-protein companies keep failing.” Business models fail; the platform did not. Insulin, rennet, and animal-free whey are all shipping. The 2024 funding downturn193 killed the companies that sold the timeline instead of the molecule. Sell the molecule.

You build it. If you would rather fund it, this page doubles as the thesis: the largest, least-optimized machine on Earth is a workaround for a problem chemistry already solved, and the first operator to make the molecule cheaper than the animal gets a planet’s worth of land, water, and health as the margin.

I love you very much and I do not want you and everyone you have ever loved to be slowly tortured and brutally murdered by horrible diseases, and I would prefer the animals not be either, but that second part is optional, and the math works without it.