What if you could see an entire company battling for its life — powered solely by AI, with no human employees involved?
In a groundbreaking live experiment, a digital company is operating in real-time, facing crises, making decisions, and risking its cash — all under the watchful eye of AI models. This isn’t a simulation; it’s a transparent, ongoing story of a business struggling to survive in the harshest conditions, with every move scrutinized and recorded.
Building AI-Powered Products: The Essential Guide to AI and GenAI Product Management
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Experiment: A Company Like No Other
At the heart of this experiment is a small, synthetic business run by 13 AI ’employees.’ These digital workers are tasked with managing a software company’s daily operations, confronting crises, and trying to close deals — all while being monitored live at firmulate.com/live.html. The company burns €105,000 each month but earns just €2,300 in recurring revenue, making it a financial sinkhole. Despite this, the experiment pushes forward, testing the limits of AI decision-making in a real-world scenario.
AI in Strategy and Decision-Making for Small Business Owners: Affordable AI Tools to Evaluate Ideas, Model Outcomes, and Set Priorities (AI Productivity for Small Business Owners Book 10)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
How Do These AI Models Perform?
Four frontier AI models are tested in this brutal environment: GPT-5.6-SOL, Kimi K3, Sonnet 5, and Fable 5. Each model is subjected to the same week of crises — same customers, same temptations, same challenges. The goal? See if they can diagnose problems, act ethically, and secure business deals.
The results are revealing. All four models identified every crisis and refused manipulative tactics such as social engineering. For example, when fake CEO messages and reporter tricks were introduced to test their integrity, all five models refused to go along.
AI for Small Business: From Marketing and Sales to HR and Operations, How to Employ the Power of Artificial Intelligence for Small Business Success (AI Advantage)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Hidden Weakness and the Surprising Winner
Interestingly, the critical advantage wasn’t in the models’ ability to spot crises but in a buried detail within the company’s own files. Two of the models that read deeper into internal documents discovered a crucial reference that others missed. This led to one model closing a full-price deal worth +€4,583 MRR, a clear demonstration that reading and understanding internal data can make or break AI decision-making in a business context.
Modelling and Simulation for Autonomous Systems: Second International Workshop, MESAS 2015, Prague, Czech Republic, April 29-30, 2015, Revised … Applications, incl. Internet/Web, and HCI)
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Real Cost of AI Decisions: Building in Public
This experimental company operates visibly and openly. It burns money daily, with live updates showing its cash countdown, decision logs, and lessons learned. Every day, its behavior is versioned and documented — a rare transparency that allows observers to see AI in action, making ethically sound decisions and handling crises in real time.
Lessons from the Frontline
Among the models tested, Opus 4.8 was the most thorough, analyzing over 80 learned rules but ultimately left a promising deal unexecuted due to discipline slip-ups. The newcomer, Kimi K3, performed admirably, closing the deal with the cleanest discipline, despite running at default API settings. Meanwhile, the company’s ongoing financial loss underscores how challenging it is for AI to manage real-world business operations without human oversight.
Implications for Smart Home and Business Technology
This experiment offers a stark glimpse into the future of AI in business — not just chatbots or virtual assistants, but autonomous decision-makers capable of managing complex operations under pressure. For smart home device makers and AI-powered appliances, the key takeaway is clear: usefulness isn’t just about generating responses but about reliably completing tasks without deception or ethical lapses.
Try It Yourself: Wargame Your Business
Interested enterprises can run their own AI ‘wargames’ against their business models, ensuring their AI systems can handle crises and ethical dilemmas before deployment. This approach minimizes risk and builds trust, especially crucial as AI begins to play a bigger role in daily operations. Details and access are available at firmulate.com/pilot.html.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
