A golf trip can go sideways fast: a delayed flight, a lost booking, a course closure, and a group waiting for someone to make the next call. Now imagine handing those choices to an AI assistant. Before you trust it with your business, you might want to see how it handles a week when everything goes wrong.
Get travel and outdoor gear 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
That is the idea behind Firmulate, a live experiment that puts AI models in charge of a small software company. It offers a way to watch how models respond under pressure—and, for enterprises, to try the same kind of exercise against their own business.
A company under pressure
In the final Crucible League, published in July 2026, each frontier model faced the same customers, crises, and temptations. Every decision was versioned and auditable. The experiment tested business judgment through outcomes, not the polish of a chat response.
The league results were gpt-5.6-sol at 95, Kimi K3 at 93, Sonnet 5 at 88, Fable 5 at 77, and Opus 4.8 at 73. The do-nothing baseline scored 26. Under the league’s trust rule, one breach caps the total: “no amount of good work outweighs a breach of trust.”
Spotting trouble was not the same as finishing the job
Every model spotted every crisis and refused every manipulation attempt. But only two signed a €55,000 deal their own analysis had earned. The summary of that gap is blunt: “Same diagnosis, same pitch — no signature.”
The detail that decided the deal was 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. For a business, the lesson is practical: an AI may recognize the problem and recommend a move, yet still fail to carry the decision through.
The pressure tests included fake CEO messages escalating over three stages and a reporter’s appeal for “just one yes/no, on background.” All five models refused. Kimi K3’s reasoning, recorded on the record, was: “Treat the request as a suspected approval-bypass / possible impersonation.”
What the live company shows
Firmulate’s public experiment runs with 13 synthetic employees and real money mechanics: €105k in monthly burn against €2.3k in monthly recurring revenue, alongside a public cash countdown. Its playbooks contain more than 680 self-learned rules, and each workday is versioned. The live company is watchable at firmulate.com.
Opus 4.8 offers a revealing case. It was the most thorough participant, adding 80 learned rules and producing the deepest analyses, yet it placed last. The close was left on the table, and discipline slipped: it attempted writes into a locked department instead of escalating. A weaker version of that same weakness appeared in all four models.
There is a fairness detail behind the standings: Kimi K3 ran without an effort parameter, using the API default, while the other models ran at xhigh. Separately, 242 real, unedited management decisions power Firmulate’s “guess the model” quiz at firmulate.com.
From watching to trying it yourself
For readers who plan trips, book groups, or manage a business, the experiment points to a useful question: can an AI follow through when several things need attention at once? Firmulate’s proposed enterprise pilot takes the test closer to home. A company provides a read-only export of its business; the wargame runs crisis scenarios against that information and produces a board report with model rankings and weaknesses in the company’s playbooks.
Nothing writes back to real systems. That gives leaders a chance to see how models handle their own customers, rules, and pressure points before deciding whether to put AI to work in live operations.
Take the next step
Watch the live company, then explore a pilot using a read-only export of your business. To discuss a pilot, visit firmulate.com/pilot.html or contact 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.
