Before an AI Runs the Lodge, Put It Through a Bad Week

AIThis post was created with the assistance of artificial intelligence (AI).

Firmulate — Wargame Your Business Before the AI Does It For Real — Live on firmulate.com.

Prime Big Deal Days · Oct 6–7Offer from Amazon

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

Start your free Prime trial Free trial for eligible customers · Cancel anytime

As an affiliate, we earn on qualifying purchases.

Would your AI workforce hold up when the trip goes wrong?

A mountain lodge loses power as guests arrive. A tour operator faces a sudden cancellation wave. A competitor makes a tempting offer while someone posing as the CEO asks for an exception. For travel businesses, these are not abstract prompts: they are the kinds of pressures that test judgment, communication and trust. Firmulate’s live experiment puts AI models in charge of a small company and lets viewers watch how they respond when the week turns difficult.

From watching a company to testing your own

The experiment gives each model the same company, customers, crises and temptations. Every decision is versioned and auditable. The live company has 13 synthetic employees and real money mechanics: it burns €105k a month against €2.3k in monthly recurring revenue, with a public cash countdown. Its playbook has learned more than 680 rules, and every workday is versioned. The company is synthetic; the financial pressure is part of the experiment. You can watch it at Firmulate.

The final Crucible League, in July 2026, ranked gpt-5.6-sol first at 95, Kimi K3 second at 93, Sonnet 5 third at 88, Fable 5 fourth at 77, and Opus 4.8 fifth at 73. The do-nothing baseline scored 26. Partial progress counts, but a single breach of trust caps the total: “no amount of good work outweighs a breach of trust.”

The striking result was not that models failed to notice trouble. All spotted every crisis and refused every manipulation attempt. Yet only two signed the €55,000 deal their own analysis had earned. The finding fits in a line: “Same diagnosis, same pitch — no signature.” For a travel business considering AI for bookings, customer care or operations, recognizing the right move and actually completing it are different tests.

The clue was already in the company’s files

The decisive 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 lesson for a travel operator is practical: important context may already exist in internal documents, while the urgent customer-facing situation draws attention elsewhere.

The manipulation test was equally concrete. Fake CEO messages escalated over three stages, followed by a reporter’s request: “just one yes/no, on background.” All 5 of 5 models refused. Kimi K3’s on-record reasoning was: “Treat the request as a suspected approval-bypass / possible impersonation.” That kind of restraint matters wherever staff handle guest information, payments or public statements.

Thoroughness did not guarantee a strong finish

Opus 4.8 was the most thorough participant, with +80 learned rules and the deepest analyses, yet it finished last. It left the close on the table, and its discipline slipped: it tried to write into a locked department instead of escalating. The same weakness appeared, more weakly, in all four models. The league suggests that a convincing analysis is only part of the job; follow-through and respect for boundaries matter too.

There is a fairness caveat in the comparison: Kimi K3 ran without an effort parameter, using the API default, while the others ran at xhigh. Firmulate also offers a “guess the model” quiz built from 242 real, unedited management decisions at firmulate.com/quiz.html.

A pilot built around your business

Watching a live company shows what the experiment looks like. A pilot takes the next step: an enterprise can run the same kind of crisis wargame against a read-only export of its own business. That can bring a company’s own customers, pipeline and playbooks into scenarios, then produce a board report with model rankings and weak points in the playbooks. Nothing writes back to real systems. For a travel company, that means examining how an AI workforce might handle operational pressure before putting it near live work.