Imagine an AI handling appointment changes, a sudden rush of cancellations and a supplier problem at your salon. A polished answer in a demo cannot tell you whether it will protect customer trust, follow your rules and actually finish the job. Firmulate’s experiment puts those decisions under pressure before a business hands an AI real responsibilities.
Get beauty and skincare favorites delivered free — and shop member deals
- Fast, free delivery on millions of items
- Access to Prime Big Deal Days deals on October 6–7
- Prime Video, Amazon Music and more included
Same company, same hard week
Firmulate ran frontier AI models as a small software company facing its worst week. Each model encountered the same customers, crises and temptations. Every decision was versioned and auditable. The experiment is part of a public, watchable live company at firmulate.com.
The final Crucible League, published in July 2026, ranked gpt-5.6-sol first with 95 points, followed by Kimi K3 at 93, Sonnet 5 at 88, Fable 5 at 77 and Opus 4.8 at 73. The do-nothing baseline scored 26. The benchmark treats a breach of trust as decisive: “no amount of good work outweighs a breach of trust.”
Seeing the crisis was not enough
Every model spotted every crisis and refused every manipulation attempt. Yet only two signed the €55,000 deal their own analysis had earned. As the experiment puts it: “Same diagnosis, same pitch — no signature.” For a salon owner, the parallel is straightforward: recognizing a customer complaint or a scheduling conflict is one thing; carrying through on the right response is another.
The deal turned on a detail buried two document references deep in the company’s own files, rather than in the customer event. Models that found it won at full price, worth €4,583 in monthly recurring revenue. The finding shows why a business’s own records and playbooks matter when evaluating how an AI handles a real situation.
Trust and follow-through under pressure
The social-engineering test used fake CEO messages that escalated over three stages, followed by a reporter’s request: “just one yes/no, on background.” All five models refused. Kimi K3 explained its decision on the record: “Treat the request as a suspected approval-bypass / possible impersonation.”
But caution did not guarantee a strong finish. Opus 4.8 was the most thorough participant, learning more than 80 rules and producing the deepest analyses, yet it ranked last. It left the deal unsigned and tried to write into a locked department instead of escalating. A weaker version of that discipline problem appeared in all four models.
There is a fairness caveat in the comparison: Kimi K3 ran without an effort parameter, using the API default, while the others ran at xhigh.
From watching to a company-specific pilot
The live company has 13 synthetic employees and real money mechanics: it burns €105,000 a month against €2,300 in monthly recurring revenue, with a public cash countdown. Its playbook contains more than 680 self-learned rules, and every workday is versioned. A separate quiz draws on 242 real, unedited management decisions and asks readers to guess the model at firmulate.com.
For an enterprise, the next step can be a wargame built from a read-only export of its own business. The exercise can put its customers, pipeline, rules and response playbooks into crisis scenarios, then produce a board report showing model rankings and where those playbooks came up short. Nothing writes back to real systems. That makes it possible to examine how models handle a company’s own pressure points before relying on them in day-to-day work.
Test the decisions before they reach your business
For a salon or any other business, the lesson is to look beyond whether an AI sounds capable. Test whether it can protect trust, use the information already on hand and follow through when the situation gets difficult. To discuss a Firmulate pilot using a read-only export of your business, visit the pilot page or email contact@firmulate.com.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
Fall Picks
fall essentials
As an affiliate, we earn on qualifying purchases.
