Susquehanna International Group, a quant behemoth, claims it lost $70 million to an insider trading scheme tied to Chinese securities options. That’s not a headline. That’s a declaration of war on information asymmetry.
The accusation, surfaced via a recent report, alleges a sophisticated cross-border operation where non-public information about Chinese equities flowed into U.S. options markets. Susquehanna, known for its high-frequency trading and market-making dominance, states it was the victim. But as someone who spent 2017 auditing ERC-20 contracts for reentrancy vulnerabilities, I know that what looks like a theft of alpha is often a battle over whose code—or whose strategy—gets to function in the open.
Let’s strip the narrative of its legal veneer. The alleged scheme mirrors the classic playbook: an insider in China, an option trade in the U.S., and a bridge made of encrypted messages and offshore accounts. But the numbers—$70 million in losses—suggest scale. This isn’t a rogue trader leaking a quarterly report. This is a structured flow of value extraction, leveraging the friction between China’s capital controls and America’s derivatives liquidity.
I’ve lived through this tension. During the DeFi Summer of 2020, I deployed €200k into Compound and Uniswap pools, dynamically rebalancing collateral to capture a 140% return in six weeks. That taught me one immutable truth: liquidity doesn’t care about jurisdiction. It flows toward the path of least resistance. If you can’t move a trade in Shanghai, you move it to Chicago. If you can’t hold a share in a Chinese A-share, you buy an option on a Hong Kong-listed ETF. The market finds its bottleneck; the insider finds its edge.
The core insight here isn’t about illegality—it’s about structural design. Susquehanna’s loss functions as a stress test for the entire cross-border derivative ecosystem. Their trading algorithms likely detected a persistent, anomalous edge in options pricing on securities correlated with Chinese indices. Statistical arbitrage that held too long, that correlated with events that hadn’t happened in public news. Every quant firm dreams of such a signal; every firm fears it because it signals someone is operating with asymmetric information.
Options don’t hedge against ignorance; they amplify it. The accused party, likely a network of traders or a single entity with access to privileged data, used the leverage of options to multiply the informational advantage. A $1 inside tip on a stock can yield $10 in return. On an option with 10x leverage, the same tip yields $100. The math is brutal. Susquehanna, as a market maker, is on the other side of every trade. They provide the liquidity, the counter-party. When the information gap is too wide, their delta-hedging fails. They bleed.
Now, the contrarian angle. Most commentary will frame this as a clear-cut crime: insider trading, bad actors, need for regulation. But from where I stand—having analyzed the Terra/Luna collapse in 2022 in real-time, watching liquidity drains at specific block heights—I see a different friction. Susquehanna is not just a victim; it’s a provocateur. By publicly accusing an unnamed scheme, they are forcing the regulatory popcorn machine to heat up. This is a strategic move, not a public service announcement.
Risk isn’t a number; it’s the gap between belief and reality. The real risk here isn’t the $70M loss; it’s the precedent of using civil litigation to map the information-supply chain of an entire geopolitical trading corridor. Susquehanna doesn’t just want money back. They want to establish a deterrent: if you trade on insider information from China using U.S. options, we will find you, and we will make the discovery process so expensive that the edge disappears. This is a signal to every quant fund operating in the cross-border space: compliance is now a competitive moat.
But there’s a blind spot. The accusation, if pursued in U.S. federal court, will run headlong into China’s data sovereignty laws. The Chinese Securities Law and the Data Security Act prohibit unauthorized cross-border transfer of important financial data. The accused party’s information source likely sits inside China’s regulatory bubble. Susquehanna’s lawyers would need to prove the chain of custody of that information—signal to trade—which requires data from Chinese servers, Chinese banks, and Chinese telecoms.
This is where the battle gets technical. In 2026, while piloting an AI-agent trading system in Paris, I saw how machine speed can outrun human regulation. The pilot processed news sentiment faster than a human could read it. But it also hallucinated trade executions. Code doesn’t cheat; it executes. A trading algorithm doesn’t care about jurisdictional boundaries. But the discovery process does. The U.S. court can order a subpoena; China can block it. The result is a legal black hole.
Arbitrage doesn’t lie; it just hides in the spread. The spread in this case is legal and regulatory. The accused party may never be named publicly. The case may settle. But the damage is done: the markets now know that quant firms are paying close attention to cross-border information flows. The cost of that attention is $70M, but the price of ignoring it was higher.
Terra’s code was poetry; Luna’s exit was prose. This situation is pure prose—messy, human, and full of intent. Susquehanna is writing a new chapter in the playbook of cross-border enforcement, one where the plaintiff functions as both market participant and intelligence agent.
The takeaway is sharp: if you trade on the edges of jurisdictions, expect the market—and its largest players—to build a mousetrap. The $70M is not a loss; it’s tuition for the entire industry on how to price geopolitical risk into every option premium. The question isn’t whether insider trading happened. The question is whether the system can detect and price it before the next $70M vanishes.