The hook is a lecture hall in Beijing. The speaker, Zhang Xiaolong, CEO of Fenbi Education, a publicly traded tutor for China's civil service exams, stands before students at Renmin University. The advertised topic: career planning in the AI age. What unfolds instead is a thirty-minute monologue on the virtues of stock speculation, a claim of personal windfall—$8.3 million in one month—and finally, an outburst at a student who dared question the narrative. He ends the lecture, leaves the stage, and resigns two weeks later. The company's stock dropped 17% on the announcement of a $8.3 million trading loss booked in its investment portfolio. The media calls it a scandal. I call it a structural failure in the architecture of trust.
Context Fenbi is not a crypto project. It is an education company—a B2C service that sells preparation courses for China’s highly competitive civil service exams. Its core product is the promise of a stable career. Its business model depends on prepaid tuition fees, which create a significant cash float. In 2021, the company listed on the Hong Kong Stock Exchange with a market cap above $3 billion. By 2026, after a bear market for education stocks, it still held over $100 million in cash. The CEO, a charismatic founder with a background in technology, had steered the company through regulatory storms and competitive pressure from incumbents like Zhonggong and Huatu. But the event at Renmin University exposed something the balance sheet could not hide: a schism between the company’s stated mission and its leader’s personal incentives.
Core: The Systematic Teardown I do not trust the pitch; I audit the structure. Here, the structure has three independent failure modes. First, capital misallocation. Fenbi’s cash was not idle—it was deployed into listed equities. The CEO actively managed this portfolio. The $8.3 million loss represents a 9.5% drawdown on reported holdings. In a pre-paid business model, this cash is not surplus; it is a liability—deferred revenue owed to students who may demand refunds. Using it for speculative trading creates a liquidity mismatch. Liquidity is a mirage; solvency is the only truth. Second, governance failure. The CEO appears to have single-handedly directed a material portion of the company's treasury into risky assets without board override or investor consent. The lecture itself was not a rogue act—it was a public manifestation of an unchecked control structure. The resignation is a symptom, not a cure. Third, communication failure. The core value proposition of Fenbi is risk reduction—a stable job via exam success. The CEO told students to speculate. This is a contradiction of the brand’s fundamental promise. In crypto, we call this a “narrative mismatch”: the protocol says one thing, the code does another. Here, the code is the CEO’s behavior.
I’ve seen this before. In 2017, I audited an ICO that claimed to be building a decentralized exchange. The whitepaper promised automated market making. The contract contained a backdoor that allowed the founder to drain any pool. The team defended it as a “security feature.” The Fenbi lecture is that backdoor—a single point of failure that voids the entire trust model. The $8.3 million loss is merely the transaction cost of discovering the flaw.
Contrarian Angle A bull case exists. Some argue the CEO’s trading was personal, not corporate—the loss was booked in an investment account separate from operating cash. They point to Fenbi’s still-solid market share and the fact that no students have yet lost access to courses. They claim the market overreacted, that the resignation will restore order, and that the company’s fundamentals remain intact. I concede two data points: Fenbi continues to generate revenue from its core business, and the pre-paid cash float is still sufficient to cover liabilities. But the bull case ignores the second-order effects. Trust is a structural variable. Once the architecture is compromised—once you prove the CEO will prioritize personal speculation over mission—the cost of customer acquisition rises, employee retention drops, and competitors gain a powerful narrative weapon. The bull case treats value as a function of current cash flows. I treat it as a function of the integrity of the system. Emotion is a variable I exclude from the equation. The math says: a broken trust structure cannot price its own risk correctly.
Takeaway Fenbi will likely survive. The CEO resigned. A professional operator will take over. But the event is not an isolated scandal—it is a case study in how quickly a central point of governance failure can destroy the equity of a business built on promise. For crypto projects, the lesson is direct: audit the structure, not the pitch. Who controls the treasury? What are the incentive alignments? Can one key holder override the system? The answer to these questions is the only forecast that matters. Solvency is the only truth. The rest is noise.