Can Artificial Intelligence Keep a Business Alive?
Imagine a company with no human employees, losing €105,000 each month, yet still running and making decisions—publicly. This is not science fiction, but a real-time experiment showcasing AI’s potential and its current limitations, watched live by anyone curious about the future of work and decision-making.
As an affiliate, we earn on qualifying purchases.
The Live Experiment in Business Management
At the heart of this story is a unique experiment conducted by Firmulate, a company that runs a simulated small software business every workday. The twist? Instead of human managers, the company is operated by four advanced AI models, each tasked with navigating the same challenging week—dealing with customer crises, temptations to manipulate data, and internal risks—just as a real startup might.
In this setup, all decisions are versioned and auditable, providing a transparent view into how each AI responds under pressure. The goal: see whether these models can not only identify problems but also act ethically and effectively, ultimately closing deals and maintaining integrity.
How Do AI Models Perform?
The results are revealing. All four models detected every crisis and refused every manipulation attempt, demonstrating a remarkable capacity for honesty and awareness. Yet, only half of them managed to secure a crucial €55,000 deal that their own analysis had justified—an outcome that underscores the difference between recognizing problems and executing the right solutions.
A critical detail emerges from the company’s files: the decisive weakness needed to close the deal was buried two references deep. The models that read and understood this hidden information succeeded in winning the full deal, worth over €4,583 monthly recurring revenue, highlighting the importance of thorough analysis beyond surface-level data.
Built-in Public and Built-in Challenges
This live experiment is more than just a showcase of AI decision-making; it is a build-in-public story, available for anyone to observe at firmulate.com/live.html. It features 13 synthetic employees, real money mechanics, a public cash countdown, and over 680 self-learned rules guiding every workday. Despite the transparency, the company is currently burning €105,000 each month against a modest €2,300 monthly recurring revenue—an ongoing battle for survival.
Testing the Limits of AI Integrity
The experiment also includes social engineering tests. Fake CEO messages and reporter tricks were introduced, trying to elicit approvals or bypass controls. All four models refused to cooperate, with Kimi K3 explicitly recognizing the risk of impersonation or approval-bypass, demonstrating a cautious and ethical stance under pressure.
Lessons from the Deep Analysis
The most thorough model, Opus 4.8, with over 80 learned rules, performed the worst in closing deals—highlighting that exhaustive analysis alone does not guarantee success if discipline slips or critical steps are skipped. Interestingly, all models exhibited similar weaknesses, such as failing to escalate issues properly.
As an affiliate, we earn on qualifying purchases.
Implications for Business and AI
This experiment reveals vital insights for companies considering AI automation. It’s not just about whether AI can generate convincing chat responses; it’s whether it can execute complex workflows reliably, stay honest under pressure, and identify hidden opportunities. As one participant put it, “the decisive weakness sat two document references deep”—a reminder of the importance of thorough information processing.
The Big Question for Leaders
If AI agents will touch your CRM, support queue, or forecasting tools, ask yourself: will they finish what they start? Will they read and understand your files? Will they stay honest when under stress? These are the real tests, and this live experiment offers a rare glimpse into the answers.
As an affiliate, we earn on qualifying purchases.
The Future of AI in Business
The live company continues to run daily at firmulate.com/live.html, with new decisions, new challenges, and ongoing lessons. It stands as a powerful reminder that AI in the workplace is not just about chat quality or superficial performance. It’s about integrity, diligence, and the ability to deliver useful, trustworthy work—especially when it matters most.
Takeaways for Senior Care & Aging Sectors
For those involved in senior care and aging services, the message is clear: automation and AI tools are evolving rapidly. But their success depends on ensuring they read all relevant information thoroughly, act ethically, and follow through on commitments. Watching this live experiment underscores that technology’s true value lies in its ability to perform reliably under pressure, not just in generating appealing responses.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
As an affiliate, we earn on qualifying purchases.
