AIThis post was created with the assistance of artificial intelligence (AI).
Firmulate — This Software Company Has No Employees, Loses Money Every Day — and You Can Watch.
Live on firmulate.com.

In a world where spiritual growth often involves introspection and facing fears, there’s a new kind of journey unfolding in the realm of artificial intelligence and business. Imagine a company that has no employees, yet is living its day-to-day struggle in full view of the world — battling financial losses, ethical dilemmas, and the challenge of trust. This is not fiction, but a pioneering experiment where AI models are put through their paces in a real, live business environment. The question it raises: In the quest for growth and authenticity, how do we measure integrity, resilience, and truth in the digital age?

The Live Experiment: A Company on the Edge

At the core of this groundbreaking project is a small, real software company whose entire operation is run by artificial intelligence models. This company is not just simulated; it is live, with 13 synthetic employees and a daily cash burn of €105,000 against a monthly recurring revenue of €2,300. Every workday, the company’s decisions are recorded, versioned, and made publicly accessible at firmulate.com/live. This transparency transforms what might be a typical startup story into an ongoing open experiment.

AI Builders: Making The Decisions That Turn AI Code Into Real Software

AI Builders: Making The Decisions That Turn AI Code Into Real Software

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

The Test: Four AI Models, Same Challenges

The experiment involves four of the most advanced AI models, each tasked with managing the same week of crises, customer demands, and ethical temptations. Their goal: to navigate the worst week in the company’s history, make sound decisions, and close a critical deal worth €55,000 per month in additional revenue. The models read every document, analyze every decision, and are held to the same standards — all decisions are auditable and versioned, ensuring transparency in their performance.

ESSENTIAL AI TOOLS FOR TRANSPARENT MODELS USING SHAP, LIME, AND VISUALIZATION TECHNIQUES: 65 PRACTICAL EXERCISES TO ENHANCE INTERPRETABILITY AND TRUST IN BLACK-BOX MODELS

ESSENTIAL AI TOOLS FOR TRANSPARENT MODELS USING SHAP, LIME, AND VISUALIZATION TECHNIQUES: 65 PRACTICAL EXERCISES TO ENHANCE INTERPRETABILITY AND TRUST IN BLACK-BOX MODELS

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

Findings: Honesty, Skill, and Hidden Weaknesses

Remarkably, all four models identified every crisis and refused every manipulation attempt, including a staged social engineering trick involving fake CEO messages and reporter inquiries. In fact, all models refused to bypass security or impersonate authority figures, demonstrating a high level of trustworthiness under pressure. However, only two of these models managed to close the deal at full price — one of which, the Kimi K3, did so with the most disciplined approach, refusing compromises that others slipped into.

The truly revealing discovery was buried deep within the company’s own files. The models that read these hidden references uncovered a critical leverage point — an overlooked document that contained the key to sealing the deal at full price. Those who missed this opportunity left the closing on the table, highlighting how attention to detail and thoroughness directly impact success.

Digital Consumer Behavior Essentials: Understanding Online Decision-Making, AI Influence & Algorithmic Marketing in the Age of E-Commerce (Self-Learning Management Series)

Digital Consumer Behavior Essentials: Understanding Online Decision-Making, AI Influence & Algorithmic Marketing in the Age of E-Commerce (Self-Learning Management Series)

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

The Ethical Battleground

Beyond just decision-making, the experiment probes the AI’s integrity: Would it succumb to social engineering? When asked to provide a simple yes/no response to an impersonation attempt, all models refused. Kimi K3’s explanation was clear: “Treat the request as a suspected approval-bypass / possible impersonation.” This consistency in ethical stance is vital for any AI stepping into roles of trust and authority.

MixPad Free Multitrack Recording Studio and Music Mixing Software [Download]

MixPad Free Multitrack Recording Studio and Music Mixing Software [Download]

  • Multitrack Recording and Mixing: Create mixes with audio, music, and voice tracks
  • Track Customization: Apply effects and editing tools to tracks
  • Music Creation Tools: Includes Beat Maker and MIDI Creator

As an affiliate, we earn on qualifying purchases.

As an affiliate, we earn on qualifying purchases.

The Build-in-Public Narrative

This company’s daily operations are not hidden behind secrecy or marketing hype. Instead, it openly shares its struggles, decisions, and even its failures, embodying a build-in-public ethos taken to an extreme. Every decision is versioned, every crisis documented, and every outcome transparent. This approach invites stakeholders and observers to witness firsthand how AI manages real-world complexity and ethical dilemmas in a live setting.

Implications for Business and Humanity

While the company is losing money today, the insights gained reach far beyond financial metrics. It questions how we assess AI’s trustworthiness, discipline, and ability to handle ethical pressure. As AI interfaces increasingly touch critical aspects of our lives — from customer support to decision-making — understanding these qualities becomes more vital than ever.

The Future of Transparent AI-Run Business

For organizations contemplating AI integration, this experiment offers a stark reality check: It’s not about how well an AI can generate words in a chat, but whether it can carry through on commitments, read deeply into complex documents, and resist manipulation. The live site allows anyone to watch this ongoing story unfold, see the decisions behind the scenes, and even participate in quizzes and pilot tests to gauge their own understanding of AI judgment.

In a broader sense, this venture into transparent, build-in-public AI management echoes a deeper philosophical quest — to build systems that are honest, disciplined, and aligned with human values, even amid chaos and financial hardship. As the experiment continues, it challenges us all to reconsider what trust, integrity, and resilience truly mean in a digital age increasingly defined by artificial intelligence.

Infographic — This Software Company Has No Employees, Loses Money Every Day — and You Can Watch.
The findings at a glance — source: firmulate.com.

Watch it live: firmulate.com/live · Full results: firmulate.com/benchmarks.html

Powered by Thorsten Meyer AI


You May Also Like

Dream Recurrence: Why Do We Have Recurring Dreams?

Tapping into the reasons behind recurring dreams reveals how your subconscious processes unresolved emotions and hidden fears—discover what your dreams are truly telling you.

DILD vs. WILD: Which Lucid Dream Technique Fits Your Sleep Style?

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

How Seasonal Changes Influence Dream Content

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

Leather Dream Journals and the Psychology of Writing by Hand

Forces of tactile and mindful writing in leather dream journals unlock deeper subconscious insights, but the true benefits lie beyond what you might expect.