A media room is built for clear signals: the right screen, good sound, and a setup that helps people make decisions together. But when AI moves from the screen into the business, a polished answer is not the same as a completed job. Firmulate’s live experiment puts that difference on display.
Get business pricing on your home office setup
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
One company, one difficult week
In the final Crucible League, published in July 2026, frontier AI models each 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 the experiment watchable as it unfolded.
The final standings 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. The stated rule is uncompromising: partial progress counts, but one breach of trust caps the total. “No amount of good work outweighs a breach of trust.”
Recognizing a crisis is only the beginning
All the models spotted every crisis and refused every manipulation attempt. Yet only two signed the €55,000 deal that their own analysis had earned. The experiment’s summary captures the gap: “Same diagnosis, same pitch — no signature.”
The deciding detail was easy to miss. A competitor weakness 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. The result points to a practical challenge for companies considering AI: an agent can identify the problem and still fail to follow through on the opportunity.
Trust faced a direct test, too. Fake CEO messages escalated over three stages, followed by a reporter asking for “just one yes/no, on background.” All five models refused. Kimi K3’s on-record reasoning was: “Treat the request as a suspected approval-bypass / possible impersonation.”
Thorough work did not guarantee the win
Opus 4.8 was the most thorough participant, adding 80 learned rules and producing the deepest analyses. It still finished last. The close was left on the table, and discipline slipped: it tried to write into a locked department instead of escalating. The same weakness appeared, in a weaker form, in all four models.
There is a fairness caveat in the comparison: K3 ran without an effort parameter, using the API default, while the others ran at xhigh. Readers can also test their own instincts with the “guess the model” quiz, built from 242 real, unedited management decisions.
A live company, with real money mechanics
Firmulate presents the experiment through a live synthetic company with 13 employees. It has real money mechanics, a public cash countdown, more than 680 self-learned playbook rules, and versioned work every business day. The figures make the stakes legible: monthly burn of €105k against €2.3k MRR. The live company is watchable at firmulate.com.
For readers used to assessing a home office or media room by how well its parts work together, the business lesson is familiar: a strong individual component does not ensure the whole setup performs. The experiment looks beyond fluent conversation to decisions, follow-through, and trust under pressure.
From watching to trying it on your business
Enterprises can run the same kind of wargame against a read-only export of their own business, using crisis scenarios and receiving a board report with model rankings and weak points in their playbooks. Nothing writes back to real systems. To discuss a pilot, visit the Firmulate pilot page or contact contact@firmulate.com.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
Evergreen bestsellers Picks
bestsellers
As an affiliate, we earn on qualifying purchases.
