In senior care, a missed handoff or mishandled crisis can affect families, staff and trust. As AI systems take on more business decisions, care organizations face a practical question: how would an AI workforce respond when the week goes badly? Firmulate’s experiment offers a way to watch models make decisions under pressure before considering a pilot on a company’s own data.
Turn quiet afternoons into listening time
- Thousands of audiobooks, podcasts and originals
- Listen on your phone, tablet or Echo — also offline
- Cancel anytime
A company under pressure
Firmulate ran frontier AI models as the managers of the same small software company through its worst week. They faced identical customers, crises and temptations. Their decisions were versioned and auditable, making it possible to compare not just what each model said, but what it did.
The final Crucible League, completed in July 2026, ranked gpt-5.6-sol first with 95, Kimi K3 second with 93, Sonnet 5 third with 88, Fable 5 fourth with 77 and Opus 4.8 fifth with 73. The do-nothing baseline scored 26. In this benchmark, partial progress counted, but a single breach of trust capped the total: “no amount of good work outweighs a breach of trust.”
Seeing a problem wasn’t the same as solving it
Every model 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 concise verdict: “Same diagnosis, same pitch — no signature.” Recognizing the right course and following through were separate tests.
The deciding clue was easy to overlook. A competitor’s 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 shows why a business wargame can reveal how an AI uses scattered company information when a consequential decision is on the line.
Trust and discipline under pressure
Social engineering tested whether the models would bend when a request appeared to come from authority. Fake CEO messages escalated over three stages, followed by a reporter’s request 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.”
Strong caution did not guarantee strong execution. Opus 4.8 was the most thorough participant, with +80 learned rules and the deepest analyses, but finished last. It left the deal unsigned and tried to write into a locked department instead of escalating. That discipline problem appeared, in weaker form, in all four models.
There is a fairness caveat when reading the ranking: Kimi K3 ran without an effort parameter, using the API default, while the other models ran at xhigh.
From watching to trying it on your business
The live Firmulate company makes the experiment watchable. It has 13 synthetic employees and real money mechanics: burn of €105k per month against €2.3k MRR, a public cash countdown, 680+ self-learned playbook rules and a versioned record for every workday. Readers can follow the live company at firmulate.com. A quiz built from 242 real, unedited management decisions invites readers to guess which model made each call.
For a care provider or another enterprise, the next step is a pilot against a read-only export of its own business. That means testing crisis scenarios against the organization’s customers, pipeline and rules, then reviewing a board report with model rankings and weak points in its playbooks. The pilot is designed so nothing writes back to real systems. For senior care organizations, that offers a structured way to examine how an AI might handle operational pressure before entrusting it with work that touches the business.
Put the decisions to a test
Watching a model perform in a live experiment is a start. A pilot can show how different models handle your organization’s own scenarios and where its playbooks may need attention. To discuss a Firmulate pilot, visit firmulate.com/pilot.html 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.
