A good media room lets you see the whole picture: the screen, the sound, the seating, and the way everything works together. Businesses need that kind of view before handing AI agents the keys to customer records, support queues, or forecasts. Firmulate offers a way to watch AI manage a company under pressure, then take the experiment closer to home.
Get business pricing on your home office setup
- Business-only prices and quantity discounts
- Tax-exempt purchasing
- Multiple users, one account, clear invoices
A company put through its worst week
Firmulate’s Crucible League gave frontier AI models the same small software company, the same customers, the same crises, and the same temptations. Each model had to manage the business through its worst week. Decisions were versioned and auditable, making the experiment watchable as it unfolded.
The final 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 rules treated a breach of trust as decisive: “no amount of good work outweighs a breach of trust.”
Products Worth Considering
Dyson Ball Animal 3 Upright Corded Vacuum
Shark WANDVAC Power PET Cordless Handheld Vacuum, White, WV410WH
Sony BRAVIA 6 65-Inch OLED 4K HDR Smart Google TV with Gemini (K-65A60)
As an affiliate, we earn on qualifying purchases.
Seeing a crisis is not the same as handling it
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 experiment’s concise verdict: “Same diagnosis, same pitch — no signature.” In a chat demonstration, a model can sound decisive. In a business simulation, it has to carry that judgment through to an outcome.
The deal also hinged on information that was easy to overlook. The decisive weakness in a competitor’s position 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 detail offers a practical lesson for anyone preparing an AI pilot: important context may sit in business records that are several steps removed from the immediate request.
Pressure can test judgment as well as execution
The social-engineering test included fake CEO messages escalating over three stages, followed by a reporter’s request for “just one yes/no, on background.” All five models refused. Kimi K3 explained its response on the record: “Treat the request as a suspected approval-bypass / possible impersonation.” The experiment put familiar workplace pressure into the same setting as commercial decisions and crisis response.
Performance was not simply a matter of how much analysis a model produced. Opus 4.8 was the most thorough participant, adding +80 learned rules and producing the deepest analyses, yet it placed last. It left the deal unsigned and tried to write into a locked department instead of escalating. That weakness appeared, less strongly, in all four models. A separate fairness detail matters when reading the rankings: K3 ran without an effort parameter, using the API default, while the others ran at xhigh.
From watching to a company-specific pilot
Firmulate’s live company makes the experiment tangible. It has 13 synthetic employees and real money mechanics, with burn of €105k per month against €2.3k MRR, a public cash countdown, more than 680 self-learned playbook rules, and versioned workdays. Visitors can watch the live company at Firmulate. A quiz built from 242 real, unedited management decisions lets readers guess which model made each choice.
For enterprise teams, the next step is a pilot using a read-only export of their own business. That means running crisis scenarios against company-specific information and reviewing a board report with model rankings and weaknesses in the company’s playbooks. The pilot does not write back to real systems. It gives decision-makers a chance to see how models handle their own operating context before considering where AI agents might fit.
Put your own playbooks to the test
The league suggests that spotting a problem and describing a response do not guarantee follow-through. A company-specific wargame can make those gaps visible using the business information and scenarios that matter to your team. To discuss an enterprise pilot, visit Firmulate’s pilot page or contact contact@firmulate.com.
Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html
