Get kitchen gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
What if your favorite kitchen gadget could run a business — and lose money every day?
Imagine a startup that operates in real time, with no employees, but with a complex, money-losing operation that you can watch unfold daily. Just like that high-end espresso machine or smart oven, this company is a living experiment in AI decision-making, deployed in the wild, facing crises and temptations, every single workday.
Products Worth Considering
KitchenAid Classic Series 4.5-Quart Tilt-Head Stand Mixer, Onyx Black
Philips 2200 Series Fully Automatic Espresso Machine with Classic Milk Frother, 2 Coffee Varieties, Intuitive Touch Display, Ceramic Grinder
Instant Pot InstantChill Ice Cream Maker with Rolled Ice Cream, No Pre-Freezing, 2-Pint
As an affiliate, we earn on qualifying purchases.
The Live Experiment: A Company as a Living Laboratory
At the heart of this story is a unique, publicly accessible experiment that turns the idea of business automation into a real-time spectacle. The company runs with 13 synthetic employees, each powered by advanced AI models. These models are tasked with managing a small software firm — complete with real cash mechanics, crises, and ethical dilemmas. Every decision they make is logged, versioned, and fully auditable, giving an unprecedented window into how AI handles complex, pressure-filled situations.
The company’s financials are stark: it burns through €105,000 each month while earning just €2,300 in monthly recurring revenue. It’s a demonstrative, live battleground for AI capabilities, with a public cash countdown that highlights just how fragile the operation is. Visitors can watch this ongoing experiment at firmulate.com/live.html.
The Big Takeaway: Trust and Capabilities in AI Decision-Making
This experiment reveals critical insights: all four AI models tested identified every crisis and refused manipulative tactics. Yet, only two managed to close a €55,000 deal. The secret wasn’t in superficial chat or surface-level responses — it lay in reading and understanding internal documents that held the company’s competitive advantage. The model that read a buried reference within internal files secured the deal at full price, demonstrating the importance of thorough information processing in AI decision-making.
Meanwhile, social engineering tactics like fake CEO messages and reporter tricks were universally rejected, showing that these models maintain integrity under pressure. The experiment underscores that the real question for AI in business isn’t just how well it can generate language but whether it can act honestly, follow through, and deliver useful, complete work under stress — qualities crucial for AI to genuinely augment or replace human decision-makers.
All these findings are accessible, transparent, and live, offering a rare look at AI’s current strengths and limitations in a real-world, high-stakes environment.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
