"Liang Wenfeng" from the UK has earned a staggering 64 billion yuan.

A new 100-billion-yuan fortune billionaire has been added to the list of people who have amassed wealth through AI.

In the 2026 UK Rich List recently released by *The Sunday Times*, Alex Gerko, known as "the Mathematician", ranks 8th with a net worth of 16.006 billion pounds (about 141.7 billion yuan).

According to the list's estimates, the net worth of this British billionaire has increased by 7.261 billion pounds (about 64.3 billion yuan) within one year. That translates to an average daily appreciation of 19.89 million pounds (about 176 million yuan) for Gerko's assets.

This massive fortune is backed by XTX, the quantitative trading company founded by Gerko. At present, XTX has ranked among the world's top market makers and is also one of the most profitable private companies in the UK. A set of per capita efficiency data can prove its strong profitability: the company currently has only 300 employees, but the per capita revenue exceeds 100 million yuan.

As a new-generation quantitative trading company, XTX and High-Flyer Quant founded by Liang Wenfeng were both established in 2015, and their heavy investment in AI computing power is equally aggressive. According to a 2024 report by the *Financial Times*, XTX's research work is completed by a supercomputer located in Iceland. To this end, XTX has spent more than 150 million pounds (over 1.3 billion yuan) on 25,000 AI chips.

The "British Liang Wenfeng" made huge profits with AI

Gerko was born in Moscow in 1979. He obtained a doctorate in mathematics from Lomonosov Moscow State University, the most prestigious university in Russia, and also earned a master's degree in economics from the private New Economic School.

With such an educational background, staying in academia seems to be the most logical path, but Gerko chose to turn to the financial market. In a report, the *Financial Times* attributed Gerko's career path choice to his setbacks in the academic field:

After obtaining his doctorate in mathematics in Moscow, Gerko came to the conclusion that he could not become a world-class intellectual.

In 2004, the frustrated Dr. Gerko joined Deutsche Bank. After that, Gerko joined GSA Capital, a British quantitative hedge fund, set up the foreign exchange trading department, and later took charge of the management of other assets. In 2015, Gerko, who had been working in the financial circle for more than 10 years, founded XTX with a small group of traders. The name "XTX" comes from the mathematical formula used in linear regression, which echoes Gerko's professional background.

XTX's business model is similar to that of well-known market makers such as Citadel Securities and Jane Street: it provides quotes to global clients and accumulates profits from a huge volume of transactions through bid-ask spreads. According to its official website, XTX uses machine learning to predict the prices of more than 53,000 financial instruments including stocks, fixed income, currencies, commodities and cryptocurrencies, and then conducts transactions based on the results.

Up to now, XTX's daily trading volume is about 250 billion US dollars. According to documents from the UK's Companies House, in 2025, the total revenue of XTX's three main UK operating entities was about 3.93 billion pounds (about 35 billion yuan).

If only looking at the revenue scale, XTX still lags far behind the top US market makers: Jane Street's net trading revenue in 2025 was about 39.6 billion US dollars (about 265 billion yuan), and XTX's revenue is less than one seventh of that. While Citadel Securities' revenue in 2025 was nearly 90 billion yuan.

But in terms of profitability efficiency, even Jane Street, which dominates Wall Street, cannot catch up with XTX. Jane Street currently has more than 3,000 employees, and based on its 2025 revenue, the company's per capita revenue is about 13.2 million US dollars (over 88 million yuan). XTX currently has about 300 employees, and its per capita revenue exceeds 100 million yuan. There is also a big premise here: Jane Street has been established for 26 years, while XTX has only been in business for 11 years.

The question arises: as a latecomer in the quantitative market, how did XTX overtake against the trend?

There are two key points: choosing the right track, and then investing heavily in equipment.

In the initial stage of its business, XTX chose the foreign exchange market where market makers had not yet achieved high penetration. This was a move tailored to local conditions: when XTX was founded, its US counterparts had already established a firm foothold in the stock and options markets, while the center of global foreign exchange trading was exactly in London. In addition, after the 2008 financial crisis, foreign exchange trading accelerated to flow from large banks to professional trading companies, and XTX caught the trend.

At the same time, Zar Amrolia, co-founder and co-CEO of XTX, used to be the global head of foreign exchange at Deutsche Bank. According to Bloomberg, less than a year after he joined, XTX surpassed Deutsche Bank in *Euromoney*'s global ranking of foreign exchange spot traders.

After choosing the right track, the next step is to deploy equipment — XTX mainly spends a lot of money to build its own AI computing power, and uses machine learning and neural network models to conduct transactions.

The story of Liang Wenfeng's High-Flyer Quant spending heavily on building AI infrastructure is well known in China: in 2020 and 2021, High-Flyer Quant successively invested nearly 1.2 billion yuan to build its own AI cluster to provide computing power support for research.

As Liang Wenfeng's peer in the UK, Gerko's investment in computing power is equally staggering. According to a 2024 report by the *Financial Times*, XTX's research work is mainly completed by a supercomputer located in Iceland. This supercomputer can store 400PB of data, which is equivalent to about 80 trillion digital photos. To achieve this goal, XTX has spent more than 150 million pounds (over 1.3 billion yuan) on 25,000 AI chips. Most of these chips are from NVIDIA, which also makes XTX one of NVIDIA's major customers.

The quantitative trading industry kicks off an arms race

The development of XTX's business has made its founder Gerko a regular on the UK Rich List.

In 2023, Gerko topped *The Sunday Times* Tax List for the first time with an estimated tax payment of 487.4 million pounds (about 4.3 billion yuan), becoming the largest taxpayer in the UK. In 2024, he once again topped the list with a tax payment equivalent to more than 6 billion yuan. In the Rich List released by *The Sunday Times* this year, Gerko became the 8th richest person in the UK with a net worth of over 140 billion yuan.

After becoming a billionaire, Gerko followed the advanced path taken by many other wealthy people and began to work as an investor and philanthropist.

According to its official website, the projects invested by XTX include popular technology companies such as Anthropic, AI chip company Groq, and autonomous driving company Wayve.

In the field of philanthropy, Gerko continues to invest funds in mathematics education and basic research. In 2023, he donated 1 million pounds to the Institut des Hautes Études Scientifiques (IHES) in France, the institution where 2026 Fields Medal winner Wang Hong works. In terms of mathematics education, Gerko and his wife also funded the establishment of 1729 Maths School. This specialized mathematics school located in north London recruits mathematical talents aged 11 to 18, and provides full or partial bursaries for students in need.

A mathematics doctor entering the quantitative investment field and devoting tens of billions of net worth to philanthropy, the labels on Gerko easily remind people of James Simons, the father of quantitative investment.

After obtaining his doctorate in mathematics at the University of California, Berkeley, Simons taught at the Massachusetts Institute of Technology and Harvard University successively. At the age of 30, he became the head of the Department of Mathematics at Stony Brook University, State University of New York. He also co-proposed the mathematical framework later known as the "Chern-Simons theory" with the famous Chinese-American mathematician Shiing-Shen Chern. In 1978, 40-year-old Simons left the academic circle and founded the quantitative investment company Renaissance Technologies.

Different from the traditional investment method that relies on analyzing company fundamentals and judging enterprise value, the quantitative investment model pioneered by Simons mainly analyzes macro data, market behaviors, corporate financial data and other information, and forms automated trading strategies through mathematical models and algorithms. When Simons passed away in 2024, his net worth reached 31.4 billion US dollars (about 210 billion yuan). According to Forbes estimates, Simons donated a total of about 6 billion US dollars (about 40 billion yuan) in his life for research in fields such as fundamental mathematics and physics, autism and neuroscience.

However, compared with the era when Simons started his business, the quantitative investment industry where Gerko works has long entered a different stage of competition.

Although today's top quantitative trading companies still regard mathematics, physics and computer talents as their core competitiveness, just like Renaissance Technologies in its early days. But after the GPT era, in addition to smart minds, in this industry that counts in microseconds and predicts the market through models, the bargaining chips for competition among top companies also include AI capabilities and computing power reserves.

After the massive investment in AI chips, XTX's investment in the AI field has extended to infrastructure. In 2025, XTX announced plans to invest more than 1 billion euros to build five data centers in Finland. The first data center is scheduled to be put into use this year, covering an area equivalent to three football fields.

XTX's peers are also increasing their investment in AI infrastructure this year.

In April 2026, Jane Street invested 1 billion US dollars in AI cloud service provider CoreWeave, and promised to spend about 6 billion US dollars in the future to use its AI cloud platform, to obtain next-generation computing power resources including NVIDIA's Vera Rubin technology. Jane Street was also exposed to be planning to build its own data centers. Other quantitative trading institutions have similar moves to increase their investment: Jump Trading already owns three large research data centers; QRT has deployed high-density computing resources through cooperation with Icelandic data center operator atNorth.

By betting on AI computing power for trading, Gerko has achieved a net worth of 100 billion yuan. On the conveyor belt of AI wealth creation, the next billionaire may already be on the way.

This article is from the WeChat official account "ChinaVenture", author: Li Xinting, published with authorization from 36Kr.