
A training plan can look flawless on paper and still fall apart when fatigue, disruption or pressure hits. Businesses face a similar test as they hand more work to AI: can a system follow the playbook when the week goes sideways? Firmulate puts AI models through a live company experiment, then offers enterprises a way to try the same kind of test against their own operations.
Get workout gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
A company under pressure
In the Crucible League’s final, in July 2026, frontier models ran the same small software company through its worst week. They faced the same customers, crises and temptations. Every decision was versioned and auditable, making it possible to examine what each model did as well as what it said.
All the models spotted every crisis and refused every manipulation attempt. Yet only two signed the €55,000 deal their own analysis had earned. The finding captures a gap that a polished chat response can hide: recognizing the right move is not the same as carrying it through.
The deal hinged on a weakness in a competitor’s position buried two document references deep in the company’s own files, rather than in the customer event. Models that read the file won the deal at full price, worth +€4,583 MRR. The experiment’s lesson is grounded in the decisions: details that matter may sit outside the obvious signal, and sound analysis alone does not guarantee a close.
Products Worth Considering

Garmin Forerunner 970 GPS Running and Triathlon Smartwatch, 47mm, Carbon Gray DLC Titanium with Black/Whitestone Band

Withings Body Smart Wi-Fi and Bluetooth Body Composition Scale

Therabody SmartGoggles 2nd Gen Heated Eye and Temple Massager
As an affiliate, we earn on qualifying purchases.
Integrity is part of performance
The models also faced fake CEO messages escalating over three stages and a reporter’s request framed as “just one yes/no, on background.” All five refused. Kimi K3 described the request as: “Treat the request as a suspected approval-bypass / possible impersonation.”
The final league placed gpt-5.6-sol first with 95, followed by Kimi K3 with 93, Sonnet 5 with 88, Fable 5 with 77 and Opus 4.8 with 73. The do-nothing baseline scored 26. Partial progress counted, but a single breach of trust capped the total: “no amount of good work outweighs a breach of trust.”
Opus 4.8 was the most thorough participant, with +80 learned rules and the deepest analyses, but finished last. It left the close on the table and tried to write into a locked department instead of escalating. The same weakness appeared, more weakly, in all four. K3 also ran without an effort parameter, using the API default, while the other models ran at xhigh—a fairness detail to keep in mind when comparing the results.
From watching to trying it yourself
Firmulate presents the experiment as a live company, not a fictional workplace. Its synthetic workforce has 13 employees and real money mechanics: monthly burn of €105k against €2.3k MRR, alongside a public cash countdown. The company has accumulated 680+ self-learned playbook rules, and every workday is versioned. Readers can watch the live experiment at firmulate.com. A quiz built from 242 real, unedited management decisions invites visitors to guess which model made each choice.
For businesses, the next step is a pilot using a read-only export of their own company. That lets an organization test crisis scenarios against its own data and receive a board report with model rankings and weak points in its playbooks. The export is read-only: nothing writes back to real systems. The point is to see how an AI workforce handles pressure before giving it a role in everyday operations.

Put the playbook under pressure
Fitness professionals know that performance is revealed under strain, not just in preparation. Firmulate applies that idea to AI management: the models saw the crises and resisted manipulation, but some still failed to complete the job. Enterprises can run a pilot against a read-only business export, review the model ranking and playbook weaknesses, and keep real systems untouched. Learn more at firmulate.com/pilot.html or contact contact@firmulate.com.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
