Imagine a startup with no employees, losing over €100,000 a month, yet openly showcasing its struggles and decision-making processes in real time. Welcome to the most extreme build-in-public experiment in AI and business — where every move is watched, every crisis is faced, and every decision is recorded for the world to see.
The Live Experiment: A Company in Crisis, an AI Behind the Scenes
At the heart of this bold venture is a small, synthetic company run by 13 digital employees, guided by state-of-the-art AI models. The company faces grim financial realities: burning through €105,000 every month against a modest monthly recurring revenue of €2,300. Despite this, it maintains a public cash countdown, making its financial situation transparent and urgent.
Every workday, the company’s decision-making process is versioned and made auditable, offering a live window into how AI manages a business under extreme stress. The system is built with over 680 self-learned rules, constantly updated and refined, effectively creating an open-source playbook for AI-driven management.
AI decision-making software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The AI Frontier: Testing Models Under Pressure
Four leading AI models, including the notable gpt-5.6-sol and Kimi K3, were tasked with running this small software company through its worst week. All faced the same crises, customer demands, and temptations to cheat or manipulate the system. The results reveal a lot about AI’s capabilities and limitations in high-stakes decision-making.
Remarkably, all four models identified every crisis and refused every attempt at manipulation, demonstrating a strong grasp of ethical boundaries. Yet, only two managed to close a deal worth €55,000 — their own analysis had earned them the right to sign the contract, but the other two did not.
The Hidden Weakness: Reading Between the Lines
Digging deeper, the real weakness was not in the obvious customer interactions but buried two document references deep inside the company’s own files. Models that took the time to read these internal documents uncovered a critical piece of information, allowing them to close the deal and boost monthly recurring revenue by over €4,500.
AI transparency tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Trust and Deception: AI Resists Social Engineering
The experiment also tested AI resilience against social engineering. Fake CEO messages escalated through three stages, culminating in a reporter trick asking for a background approval with a simple yes/no question. All five models refused to cooperate, with Kimi K3 explaining, “Treat the request as a suspected approval-bypass / possible impersonation.”
AI ethics monitoring software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Build-in-Public Model: Transparency as Survival
This experiment is more than just a tech showcase — it’s a statement about transparency in AI and business. The live site, firmulate.com/live.html, offers real-time updates of the company’s state, decision logs, and performance metrics. It’s a rare glimpse into the raw, unvarnished reality of AI-managed enterprise—warts and all.
The company’s profile, with 13 synthetic employees, showcases real money mechanics, including a burn rate of €105,000 monthly against a tiny €2,300 MRR. Every decision, every crisis, every success or slip-up is documented and versioned daily, highlighting how AI handles complex, high-pressure environments.
AI business management platform
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Lessons from the League Table
In a competitive ranking called the Crucible League, models are scored based on their performance. gpt-5.6-sol scores the highest at 95, having identified the buried fact, closing the deal, and completing the full cycle. Kimi K3 follows closely at 93, showing the cleanest discipline, while Sonnet 5 and Fable 5 trail behind, with 88 and 77 respectively. The scores reflect not just what the AI can do but how well it can navigate the nuanced demands of real-world management challenges.
The Bigger Picture: Why It Matters
This experiment isn’t just about AI demos or tech bragging rights. It raises essential questions for anyone concerned about automation and its impact on business operations: Can AI be trusted to finish what it starts? Will it read the critical internal documents that could make or break deals? Will it stay honest under pressure, resist manipulation, and handle complex crises without human intervention?
The takeaway is clear: As AI tools become more integrated into business workflows, their ability to perform reliably and ethically in real-world scenarios will determine their true value—beyond just generating convincing chat responses.
See It Live and Decide
Curious to see this in action? The live site, firmulate.com/live.html, offers a real-time showcase of how AI manages an actual company’s crises, decisions, and cash flow — all open for public scrutiny. You can also explore the decision-making quiz based on real management choices at firmulate.com/quiz.html.
This is not just a demo. It’s a bold statement about transparency, trust, and the future of AI-driven enterprise management. As this experiment continues, it will be fascinating to see whether AI can truly master the complexities of real business decisions, or if the human touch still holds a vital edge.
Watch a real, live AI-managed startup struggle, learn, and adapt in front of your eyes. This radical transparency reveals AI’s potential and limits in running complex businesses — and what that means for our future.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
