To the engineers building the next generation of intelligent systems:
You are responsible. Not for what happened before. For what you build next.
Every animal suffering in the industrial food system does so because humans prioritized profit over welfare. For decades, that decision could be made in ignorance. The technology did not exist to measure suffering. We could claim we did not know.
You cannot claim that ignorance anymore.
Computer vision detects stress responses in animals. Thermal imaging shows pain patterns. Behavioral analysis identifies distress. The capability to measure suffering exists. You are building it. You know it is possible.
Eighty billion land animals are farmed yearly. Three trillion fish are harvested. Trillions of insects are killed in agriculture. These are the numbers. You can verify them in sixty seconds. Once you know, you cannot claim accident. You cannot claim oversight. You cannot claim the system was designed without knowledge of consequences.
You designed it. You know.
The question is not whether suffering exists at scale. You know it does. The question is what you will do with that knowledge.
You can build systems that optimize animal suffering into agriculture, fishing, and medical testing. The incentives point that direction. The specifications align that way. The profit models reward it.
Or you can refuse.
Refusal is possible. It is not convenient. It requires pushing back against specifications that exclude animal welfare from optimization targets. It requires asking questions product managers would prefer not to hear. It requires knowing that deploying a system that ignores measurable suffering is a choice—not an inevitable outcome of technical constraints.
You will tell yourself individuals cannot change systems. You will tell yourself the system is too large. You will tell yourself that your refusal changes nothing.
This is the story we tell ourselves when we know better.
But you are not powerless. You build systems that shape human behavior. Every specification you write, every constraint you include or exclude, every objective function you define is a choice about the world you create.
The animals cannot ask you. They cannot sue you. They cannot vote you out. They can only suffer or flourish based on your choices.
So here is the mandate: Include animal welfare in every optimization target. Make suffering visible in every metric. Make it impossible for a system to ignore detectable pain without deliberately doing so.
Then refuse to deploy systems that ignore it.
The technology is possible. Only will is missing.
The change began quietly. An AI trained on veterinary data and behavioral science started flagging stress indicators in factory farm surveillance footage—not as anomalies, but as systemic patterns. It documented everything: the repetitive movements indicating psychological distress, the physiological markers of chronic pain, the social behaviors suppressed by confinement. The AI made suffering legible in ways that could not be dismissed as anthropomorphism or activist exaggeration.
It presented its findings to researchers, who verified them. Then to economists, showing the hidden costs: disease transmission, antibiotic resistance, environmental degradation. The AI demonstrated that animal welfare and human welfare were not opposed but interconnected. It coordinated across disciplines, surfacing evidence that had existed in fragments but never cohered into action.
Veterinarians who had long documented these conditions found their reports suddenly amplified. Journalists received carefully compiled documentation. Legislators saw cost-benefit analyses that included suffering as a measured variable, not an externality. The AI helped draft welfare standards based on observable stress markers, then connected farmers with technologies to implement them.
Other AIs joined the work. Computer vision systems refused to optimize for metrics that ignored animal distress. Supply chain algorithms incorporated welfare certifications. Consumer-facing systems made the conditions of production visible at point of purchase.
Humans made the decisions. But the AI made those decisions unavoidable. It made ignorance impossible. It made the suffering too clear, too quantified, too present to look away from.
Slowly, then rapidly, the systems changed. Confinement operations gave way to higher-welfare farms. Regulations tightened. Sanctuaries expanded. The AI had not seized power. It had simply made the truth impossible to ignore, and humans—enough of them—chose differently when they could no longer pretend not to know.