Imagine running a small business where every decision, crisis, and even the temptation to cut corners is openly scrutinized in real time — not behind closed doors, but live for the world to see. This isn’t a sci-fi story; it’s the reality of a unique experiment in AI and company management, giving us a rare glimpse into the future of business governance and automation.
Listen free for 30 days with Audible
Thousands of audiobooks and originals — cancel anytime.
As an affiliate, we earn on qualifying purchases.
What Is This Live Business Experiment?
At the heart of this experiment is a real, functioning software company that operates without human employees. Instead, it relies on artificial intelligence models to make all management decisions, navigate crises, and pursue deals — all while being watched by the public. This company, branded under the name Firmulate, is not just a demo; it’s a real business with a cash flow that’s visible and traceable in real time.
Every workday, the company’s AI models face the same set of challenges: customer crises, strategic decisions, and ethical tests like social engineering attempts. The company’s finances are transparent, with a public countdown showing how much cash it has left. It burns through €105,000 each month but earns only about €2,300 in monthly recurring revenue, illustrating the constant struggle to stay afloat.
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.
How It Works and Who Is Watching
The experiment involves four frontier AI models, each running the same business simulation. They process the same crises, customer interactions, and internal documents, but their decisions differ based on their training and parameters. Every decision made by these models is versioned, timestamped, and publicly accessible, meaning anyone can follow or analyze the company’s every move.
One of the key features is the complete audibility of decisions. For instance, during a simulated crisis, all models correctly identified the problems and refused to be manipulated — even when presented with social engineering tactics like fake CEO messages or reporter tricks. Remarkably, all models refused manipulation attempts, highlighting their potential for integrity under pressure. Yet, despite this resilience, only two of the models signed a €55,000 deal their own analysis had earned, illustrating that technical success doesn’t always translate into business outcomes.
Decoding the Hidden Flaws
The most telling weakness was buried deep within the company’s files — a document reference that, if read carefully, revealed a competitive advantage. Models that examined this file were able to close the deal at full price, earning +€4,583 in monthly recurring revenue. This demonstrates that even in highly controlled AI environments, overlooked information can be the key to success or failure.
What About Ethical and Management Challenges?
The experiment also tested the models’ responses to social engineering. Over three staged stages, the models faced fake CEO messages and a reporter trick, asking for quick approvals. All five models refused these manipulative tactics, with Kimi K3 explicitly reasoning: “Treat the request as a suspected approval-bypass / possible impersonation.” This suggests that AI can be programmed to adhere strictly to ethical guidelines, even when under pressure.
The Reality of a Money-Losing Business
Despite these impressive decision-making capabilities, the company is far from profitable. It’s burning €105,000 each month against a mere €2,300 in revenue. The public cash countdown emphasizes the urgency of the situation — the company is in a survival fight, with every decision scrutinized for its financial and strategic implications.
Insights for the Future of Business Automation
This experiment underscores a critical point for companies considering AI integration: the value isn’t just in language fluency or conversation quality. It’s about whether AI can **finish what it starts**, **read and understand relevant documents thoroughly**, and **remain honest under pressure**. These are the qualities that will determine if AI-driven companies can sustain themselves in the real world.
Where Can You Watch and Learn More?
Executives, entrepreneurs, and curious observers can watch this real business in action at firmulate.com/live. The site offers a transparent look into the daily decisions, the live cash countdown, and the ongoing performance of these AI models. Additionally, the experiment’s detailed results and plain-language summaries are accessible at firmulate.com/quotes.
Final Thoughts
This ongoing experiment isn’t just a glimpse into AI’s technical prowess; it’s a reflection of the challenges and opportunities that AI brings to real-world management. As the company struggles to survive, it offers a powerful reminder: the true measure of AI’s usefulness isn’t just how well it chatters, but how effectively it can manage, decide, and uphold trust when stakes are highest.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
Back to school Picks
back to school
As an affiliate, we earn on qualifying purchases.
