At the start of 2012, a portfolio held by JPMorgan Chase’s Chief Investment Office kept breaching its risk limits. Then a new value-at-risk model arrived, and the problem seemed to vanish. Under the new model, the portfolio’s reported risk fell by roughly half.
Months afterward, the trades that came to be called the “London Whale” had lost more than $6 billion. When the bank investigated, its own review found that the model ran on spreadsheets fed by manual copy-and-paste steps, and that its calculations contained errors.
The trades themselves were dangerous. Far more unsettling was the figure that called them safe.
This contradiction sits at the heart of algorithmic trading risk management. Every risk system is also a model: designed by people, written in software, and capable of error. Institutions that survive have learned to ask a question most traders never consider: who measures the measurer?
Why VaR Is Not Enough
Value at risk won its popularity by reducing a portfolio’s danger to a single figure. That simplicity is also its weakness. It describes what losses look like on ordinary days and says almost nothing about how bad the worst days can get.
Regulators absorbed the lesson. The Basel Committee’s revised market risk framework moved bank capital models from VaR to expected shortfall, a measure that asks what the average loss looks like once you are already in the tail.
The principle applies far beyond banking. Serious quantitative trading research treats risk metrics with the same skepticism it applies to alpha signals. Every metric is tested against historical crises, its assumptions are written down, and changes to its methodology go through independent review. An unchallengeable risk model is just an opinion with decimals.
Constrain How You Trade, Not Only What You Hold
The Flash Crash struck U.S. markets on May 6, 2010. According to the joint SEC and CFTC report, a key trigger was a sell program of 75,000 E-mini S&P 500 futures contracts. The algorithm executing it targeted a share of trading volume without regard to price or time.
The position was not exotic. The execution was unconstrained.
Institutions now write execution limits directly into AI trading system development. These include price collars that block orders far from the prevailing market, participation caps, order-rate throttles, and checks that pause an algorithm when conditions change faster than it was designed to handle. Exposure limits tell you how much you can lose. Execution limits decide how fast.
Layer the Defenses
Mature risk architectures rarely rely on one control. They stack them.
Pre-trade checks reject orders that violate size, price, or credit limits before those orders leave the firm. Intraday monitoring tracks positions and losses in real time, with hard limits that halt trading and soft limits that trigger escalation. Post-trade review reconciles what happened against what was supposed to happen.
Each layer assumes the one before it might fail. That assumption is not pessimism. It is engineering.
Risk Logic Deserves Production Rigor
Ordinary software problems contributed to the London Whale model’s failure: manual processes, too little testing, and weak change control. Institutional algorithmic trading software development now treats risk logic with the rigor usually reserved for trading engines. Changes are version-controlled and peer-reviewed, and a new model runs in parallel with the old one before it takes over.
Firms also run historical crises back through their systems to confirm that controls actually fire. Here AI trading infrastructure consulting often adds the most value, since an outside team with no attachment to the existing design can test whether safeguards hold under conditions their builders never imagined.
When a Limit Breaks
A limit is only as strong as the response when it is crossed. In disciplined institutions, a breach triggers escalation to people independent of the trading desk, and the default response is to reduce risk, not to revisit the limit.
The riskiest moment in any risk framework comes when read more someone explains a breach away.
The Four-Question Audit
Before you trust a risk framework, ask four questions:
Was the risk model validated by people independent of its owners?
What does the system do in the tail, not just on an average day?
Can an algorithm lose money faster than a human can intervene?
When a limit is breached, which changes: the portfolio or the limit?
Why Risk Numbers Need Doubt
Risk numbers hold an odd emotional power. They help people sleep and let committees approve. They make uncertainty look like knowledge.
Institutions surround these numbers with doubt for exactly that reason.
The London Whale trades did not hide. They sat in plain view and were measured every day. The danger was never invisible. It was recorded in a number everyone wanted to believe.