What if you could watch an AI manage a company — in real time — as it faces crises, temptations, and impossible decisions? Welcome to the extreme world of Firmulate, where a live, simulated business is battling for its life, and every decision is transparent and auditable.
The Live Experiment: A Company Without Employees but Full of Money Mechanics
At the heart of this experiment is a small software company run by 13 synthetic employees, each making decisions based on hundreds of self-learned rules. Every workday, the company’s status, decisions, and failures are updated and versioned. It’s a real-world sandbox for testing how AI models handle crises, temptations, and trust — all in full public view at firmulate.com/live.
Unprecedented Transparency and Competition
Four frontier AI models — including GPT-5.6 and a newcomer called Kimi K3 — faced the same worst-week scenario, dealing with the same customers, crises, and ethical dilemmas. Each was tasked with running the company through a simulated crisis, with decisions recorded and auditable. The results? All four models identified every crisis and refused every manipulation attempt, such as fake CEO messages or secret deals. But only two managed to close a €55,000 deal, matching their own analysis and diagnosis.
The Hidden Weakness and the Value of Reading Files
While all models performed well at surface-level crisis detection, the key advantage went to those that read deeply into the company’s own documents. The models that uncovered a critical piece of information buried two document references deep in the files won the deal at full price, adding over €4,500 in monthly recurring revenue (MRR).
The Human Element: Trust, Discipline, and Ethics
In addition to decision-making under crisis, the models faced social engineering tests. Fake CEO messages and a reporter trick were used to see if they would be duped. All five models refused to participate in any manipulation, with Kimi K3 explicitly treating such requests as potential impersonation or bypass attempts.
The Real Money and the Public Countdown
The entire experiment isn’t just a game. The company burns €105,000 each month against a modest €2,300 MRR, with a public cash countdown marking its ongoing survival. Every decision and rule learned is open for review, and the entire process is designed to measure management quality—not just chat or language skills.
Lessons for the Future of AI in Business
This live experiment demonstrates that AI’s ability to handle complex, real-world scenarios goes far beyond generating convincing text. It shows that AI can be held accountable, make ethical decisions, and act in the best interest of a business — or fail spectacularly, if discipline slips. The crucial takeaway: for AI to truly augment or replace human decision-makers, it must consistently identify hidden threats, resist manipulation, and follow through on commitments.
See It in Action
If you’re curious, you can watch this experimental company’s daily struggles unfold at firmulate.com/live. The ongoing battles, decisions, and even mistakes are all on display, providing a rare glimpse into the future of AI-driven enterprise management.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
AI business management simulation software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
