Imagine your gym’s management team facing a week of crises—demanding customers, tempting shortcuts, and tough choices. Now, picture replacing your team with AI models, each with a distinct personality. Would they handle it better? That’s exactly what a groundbreaking experiment by Firmulate puts to the test.
The Live AI Business Wargame: Putting AI Models to the Test
In a real-world simulation, four frontier AI models were tasked with running a small software company through its worst week. The company faced the usual chaos: customer complaints, financial pressure, and ethical dilemmas—mirroring the unpredictable environment of a busy gym or fitness center. Every decision was observable, every crisis identical, and every temptation to cheat or cut corners carefully monitored.
What makes this experiment unique is that all decisions made by these AI models were real, unedited, and auditable. They had to navigate the same crises, recognize the same opportunities, and uphold the same standards. The question was simple: could these AI models perform management tasks as well as, or better than, humans? The results were revealing.
AI management decision-making tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Key Findings From the Experiment
- All four models identified every crisis and refused manipulation attempts, demonstrating a strong capacity for integrity under pressure.
- Only two models signed off on a €55,000 deal that their own analysis justified—a critical business decision—highlighting different management personalities.
- Interestingly, the decisive edge didn’t come from reading customer emails but from deeper document analysis within the company’s files. The models that read two document references deep in the company’s records successfully closed the full-price deal, adding €4,583 monthly recurring revenue (MRR).
- When fake CEO messages were escalated in staged social engineering attacks, all five models refused to approve any suspicious requests, citing concerns about impersonation or approval bypass. This demonstrates a built-in caution and ethical stance.
As an affiliate, we earn on qualifying purchases.
The Personalities Behind the Models
The experiment also highlighted the different ‘personalities’ of these AI managers:
- The gpt-5.6-sol scored highest overall, with a 95 out of 100, and managed to close the deal by uncovering the hidden documentation and maintaining integrity.
- The Kimi K3 (a newcomer scoring 93) ran without extra effort parameters and exhibited the cleanest discipline, also closing the deal at full price.
- The Sonnet 5 scored 88, closing the deal but with a few process slips, showing a slightly more relaxed management style.
- The Opus 4.8, despite being the most thorough with over 80 learned rules and deep analyses, scored 73 and left the close on the table, illustrating that being overly cautious can sometimes hinder decisive action.
As an affiliate, we earn on qualifying purchases.
What This Means for Your Business
The real takeaway isn’t just whether these AI models can make decisions—it’s how their personalities influence outcomes, especially under pressure. In a gym or fitness business, where customer trust and operational discipline are vital, understanding how these AI managers behave can inform smarter automation choices.
For example, an AI that prioritizes integrity and thorough analysis (like gpt-5.6 or Kimi K3) is more likely to avoid shortcuts and unethical decisions, safeguarding your reputation. Conversely, more relaxed models might be faster but risk slipping into process slips, potentially damaging trust over time.
As an affiliate, we earn on qualifying purchases.
Experience the Live Business Simulation
Curious to see how these AI models perform in your own business context? You can run a similar scenario against your company’s data in a safe, read-only environment. This real-time wargame helps you gauge whether your AI workforce will finish what it starts, read your critical files, and stay honest under pressure—before you deploy it in production.
Visit firmulate.com/quiz.html to test your management decisions and discover which AI personality best suits your operational style. It’s an engaging, transparent way to assess AI readiness for your business challenges.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
