What if an AI could run a business — and you could watch it struggle in real time?
Imagine a company with no human employees, burning through €105,000 every month while earning just €2,300 in recurring revenue. This is not science fiction but the real-time experiment of a live AI-driven company, now available for public viewing. It’s a ground-breaking attempt to test how AI manages crises, makes decisions, and stays honest under pressure — all in front of your eyes.
AI decision-making software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Live Experiment: A Company in Crisis
The setup is both simple and extraordinary. The company is run by 13 synthetic employees, each modeled by advanced AI systems. These models are put through a simulated week filled with real-world business crises — customer complaints, market fluctuations, and internal temptations to cut corners. Every decision they make is versioned, auditable, and scrutinized in real time.
The experiment’s goal: see if these AI models can manage a small software business, make honest decisions, and close deals in a challenging environment. The results are revealing. All four models tested managed to spot every crisis and refused every attempt at manipulation, such as fake CEO messages or background approval requests. Only two of them managed to close a €55,000 deal, which their own analysis identified as the right choice. The other two missed this opportunity, highlighting that even the best AI can stumble without thorough information processing.
The Hidden Weakness: Reading Beyond the Surface
Interestingly, the decisive factor wasn’t just the AI’s ability to diagnose problems but its capacity to read deep into the company’s own files. The models that examined internal documentation uncovered a critical detail buried two document references deep. This insight led to closing the deal at full price, adding over €4,500 to the monthly recurring revenue.
business AI simulation tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Building Public, Building Under Pressure
This experiment is as much about transparency as it is about testing AI. The company operates publicly, with every workday versioned, every decision logged, and a live stream available at firmulate.com/live. Over 680 rules have been learned by the AI models themselves, shaping their strategies and responses. The goal: understand how AI handles real-world business pressures without human intervention.
Not all models perform equally. For instance, Opus 4.8, the most thorough participant with over 80 learned rules, placed last in closing deals. It left potential deals unexecuted, demonstrating that even deep analysis doesn’t guarantee success if discipline slips. Meanwhile, Kimi K3, running without a default effort parameter, showed the best discipline and managed to close deals, albeit with less comprehensive rules.
Building in Public — With a Purpose
Why does this matter? Because it’s not just about AI chatbots or customer support. It’s about AI taking on real roles within your business — whether in sales, support, or management — and whether it can be trusted to act honestly and effectively under stress. This live experiment reveals that AI can recognize crises, reject manipulative tactics, and make disciplined decisions — but only if it’s built and tested rigorously.
AI project management tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Implications for Your Business
For companies in home energy, solar, or backup power, the question isn’t whether AI can generate nice-looking reports or handle simple tasks. It’s whether AI can truly manage complex, high-stakes decisions, stay honest in challenging scenarios, and deliver measurable value. The experiment shows that AI’s ability to read and interpret internal documentation was the critical factor in closing a deal — a lesson that applies across industries.
As AI becomes more integrated into business operations, watching this live company offers valuable lessons. It’s about managing AI’s weaknesses, leveraging its strengths, and understanding what it takes to build trustworthy, effective AI systems.
AI crisis management software
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Takeaway
This live AI company experiment demonstrates that the path to trustworthy automation isn’t just about chat quality. It’s about how AI handles crises, reads internal data, and stays disciplined under pressure — critical factors for any business considering AI integration today.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
