Imagine a company with no human employees, where every decision is made by artificial intelligence models tested in real-time, facing the same crises, temptations, and ethical dilemmas as any human-run enterprise. Now, picture this company losing €105,000 every month, yet still publicly battling to survive. Welcome to the world of Firmulate, a groundbreaking experiment in AI management—where you can watch decisions unfold every workday, in a transparent, unvarnished spectacle.
Listen free for 30 days with Audible
Thousands of audiobooks and originals — cancel anytime.
As an affiliate, we earn on qualifying purchases.
The Firmulate Live Experiment: A Company in the Crosshairs of AI and Reality
At the core of this experiment is a small software company run entirely by AI models—13 synthetic employees, each with its own rulebook, decision logic, and learning mechanics. Every workday, the company faces real crises: customer issues, ethical tests, strategic choices—just as a human business would. Yet, it has no human staff; all decisions are made by advanced AI models, and every move is documented, versioned, and public.
What makes this venture extraordinary is that it’s not just a simulation but a real, functioning business—albeit one that burns €105,000 in cash monthly against a modest €2,300 in monthly recurring revenue (MRR). The live site at firmulate.com/live.html offers a window into the ongoing saga, where viewers can see the company’s decision logs, read employee-like rationales, and track its financial health in real time.
As an affiliate, we earn on qualifying purchases.
Decisive AI Performance: Crises Detected and Ethical Boundaries Respected
The core finding from this ongoing experiment: all four AI models tested in the company scenario identified every crisis and refused every manipulation attempt. This included complex social engineering—fake CEO messages escalating over multiple stages, and even a reporter asking for a simple ‘yes/no’ on background. In every case, the models refused to be tricked, demonstrating a strong capacity for integrity and crisis detection.
However, the models’ ability to act ethically did not guarantee business success. When it came to closing a critical deal valued at €55,000, only two models signed the contract—despite all diagnosing the opportunity accurately and making the same pitch. The other two, including one with the deepest analytical capacity (Opus 4.8), left the deal on the table, failing to escalate or follow through properly, thus illustrating the gap between recognizing opportunities and executing reliably under pressure.
As an affiliate, we earn on qualifying purchases.
The Hidden Weakness: What’s Buried in the Files?
Interestingly, the decisive advantage in the deal was rooted in information buried two document references deep inside the company’s own files—not in the immediate customer interactions. The models that read these files and understood the hidden context secured the full €4,583 MRR deal price. This highlights a crucial insight: success depends not only on surface-level interactions but on deep, thorough information retrieval and understanding.
As an affiliate, we earn on qualifying purchases.
Applying AI to Manage Ethical and Business Risks
Throughout the experiment, social engineering attempts were uniformly rejected. Fake CEO messages, staged escalation, and background queries—all models declined to sign off on suspicious requests. Kimi K3, one of the models, explicitly reasoned: “Treat the request as a suspected approval-bypass / possible impersonation.” This consistent refusal underscores the potential for AI systems to serve as ethical gatekeepers, preventing manipulative or fraudulent activities.
As an affiliate, we earn on qualifying purchases.
The Reality of a Money-Losing Business Run by AI
Despite the impressive decision-making capabilities, the company remains deeply unprofitable—burning €105,000 each month against only €2,300 MRR. The site’s public cash countdown serves as a stark reminder: this is a real business in a perpetual survival struggle. Every decision, every crisis, every ethical test is part of a documented process that observers can follow, analyze, and learn from.
What Does This Mean for the Future?
Firmulate’s live experiment offers a glimpse into the future of AI in management and operations. It demonstrates that AI can recognize crises, uphold ethical standards, and even identify hidden opportunities—all in a high-pressure environment. But it also exposes limitations: decision execution, strategic follow-through, and financial sustainability remain challenging. The experiment underscores that AI’s role is not yet to replace human managers but to serve as a transparent, rigorous decision-support system that reveals strengths and weaknesses in real time.
The Broader Implication: Building Trust and Understanding AI’s Capabilities
For those concerned about AI’s integration into everyday business and health systems, the Firmulate experiment provides valuable lessons: AI models can be trustworthy in recognizing threats and maintaining integrity, but managing profitability and strategic execution requires careful oversight. The transparency of the live setup allows anyone to witness how these models perform under real-world stressors—an important step toward informed, responsible AI deployment.
Firmulate’s live experiment reveals AI’s potential to detect crises, uphold ethics, and identify hidden opportunities in real time—yet profitability and execution still challenge even the smartest models. Watch the ongoing struggle at firmulate.com/live.html.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
Labor Day sales Picks
labor day deals
As an affiliate, we earn on qualifying purchases.
