Imagine a company with no human employees, losing €105,000 every month, yet still fighting to close a €55,000 deal — all watched live as AI models make decisions under real crises. This is not science fiction; it’s the groundbreaking experiment from Firmulate, revealing how AI performs in the harshest management test imaginable.
The Unfolding Business Drama
At the heart of this experiment is a live simulation where 13 synthetic employees run a small software company facing its worst week. Every decision, crisis, and temptation is meticulously timed and versioned, revealing how different AI models navigate complex management scenarios. The goal? To assess whether these models can not only identify problems but also act ethically and decisively under pressure.
![Express Schedule Free Employee Scheduling Software [PC/Mac Download]](https://m.media-amazon.com/images/I/41yvuCFIVfS._SL500_.jpg)
Express Schedule Free Employee Scheduling Software [PC/Mac Download]
Simple shift planning via an easy drag & drop interface
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Key Findings from the AI Managers
Four frontier AI models were tested, each running the same challenging week. Remarkably, all four spotted every crisis — from customer complaints to operational threats — and refused every manipulation attempt, including social engineering tricks like fake CEO messages. Yet, only two of these models managed to close the €55,000 deal based on their own analysis, demonstrating a crucial gap between diagnosis and execution.
Interestingly, the decisive factor wasn’t in the immediate customer interactions or superficial decision-making. Instead, the winning models identified a buried fact buried two document references deep in the company’s files — a critical piece of information that ultimately justified the full-priced deal worth +€4,583 monthly recurring revenue (MRR). This highlights how reading and understanding internal documents can be game-changing, even more than surface-level cues.
internal document reading AI tools
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Trust, Ethics, and AI in Management
The experiment also simulated social engineering — the kind of manipulation tactics used by real-world fraudsters or unethical actors. All five models refused to be manipulated, with one model explicitly treating the fake CEO message as a suspected impersonation. This underscores that the AI models demonstrated a strong capacity for ethical decision-making, even under stress.
AI-Powered Cybersecurity: AI Tools for Enterprise Security | AI for Network Security | AI Risk Management | AI in Cyber Policies | Cyber Threat Management AI | ML in Fraud Prevention
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
The Build-in-Public Business Model
This live experiment is part of a bold, build-in-public approach by Firmulate. The company operates with 13 synthetic employees, burning €105k monthly against just €2.3k MRR, with its cash countdown publicly visible. Every workday, the system is versioned and updated, creating a transparent window into the ongoing struggle of managing an AI-run company.
Artificial Intelligence-Based Energy Management Systems for Smart Microgrids
As an affiliate, we earn on qualifying purchases.
As an affiliate, we earn on qualifying purchases.
Lessons for Industries Like Solar & Energy
For readers involved in solar energy, home backup systems, or energy management, the core lessons are clear. As AI begins to touch critical infrastructure and customer-facing processes, the questions are not just about AI’s ability to generate content or perform tasks — but whether it can finish what it starts, read crucial internal data, and operate ethically under pressure. Trustworthiness and decision integrity become vital, especially when safety, compliance, and financial outcomes are at stake.
What This Means for Your Energy Business
Just as the experiment reveals the gaps in AI’s ability to close deals or avoid manipulation, energy companies must consider how AI will handle system management, support, and customer relations. Will your AI agents read the internal documentation necessary for safe operation? Will they resist manipulation attempts? Can they deliver consistent, trustworthy work day after day?
Join the Experiment
Firmulate offers enterprise clients the chance to run similar wargames on their own business, without risking real systems or data. These simulations can prepare your AI tools to handle crises, ethical dilemmas, and complex decision-making — critical for industries relying on trust and reliability.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
