Cover Story: Trading by Algorithm: Who is Responsible When AI Calls the Shots?
00:00/00:00
Listen to this article
1x
Subscribe now
Listen to the full version
Listen to the full version
It was a battle unlike anything Wall Street had ever seen. No hedge fund managers. No financial analysts. No humans at all. Just lines and lines of code.
In the final weeks of 2025, a group of the world’s most powerful AI models — some from Silicon Valley giants, others born in the black boxes of Chinese quant funds — were given $10,000 each and released into the real-world U.S. stock market.
You've accessed an article available only to subscribers
VIEW OPTIONS
Unlock exclusive discounts with a Caixin group subscription — ideal for teams and organizations.
Subscribe to both Caixin Global and The Wall Street Journal — for the price of one.
- RELATED
Share this article
Open WeChat and scan the QR code
- DIGEST HUB
Digest Hub
Back
What Happened When
- March 2020：
- S&P 500 triggered four circuit breakers in 10 days due to COVID-19 volatility.
- 2024：
- US SEC brought its first AI-related enforcement action, fining two investment advisers for exaggerating their AI capabilities.
- In late 2024：
- China’s first LLM-constructed index series, the 'High Degree Index,' was launched by Gao Hua Securities, Sino-Securities Index Information, and Baidu Cloud.
- Oct. 8, 2024：
- A model spotted a triple-layer signal during a market downturn and cut exposure early.
- One year after late 2024：
- The flagship High Degree Index posted a 23.2% return.
- Early 2025：
- Semiconductor rally in China led to AI strategies piling into the same stocks, triggering concerns over homogeneity.
- March 2025：
- The International Organization of Securities Commissions highlighted vendor concentration risk in AI.
- By 2025：
- China’s quant industry managed around 1.4 trillion yuan in assets, with nearly 30% of strategies incorporating AI.
- 2025：
- China’s first AI investment advisory guidelines issued in Guangzhou, emphasizing human-AI collaboration.
- 2025：
- Japan’s Financial Services Agency issued warnings about over-dependence on AI in investment management.
- 2025：
- India required AI-using firms in trading/advisory to establish governance and stress-test models.
- October 2025：
- Simuwang.com data showed large-scale Chinese quant funds delivered 44.65% average return from long-only strategies.
- Early November 2025：
- ByteDance’s Seed led briefly in the RockFlow “RockAlpha” competition with a 7.09% return.
- Nov. 3, 2025：
- Alpha Arena 1.0 crypto investment tournament (in which Alibaba’s Qwen3-Max won) ended.
- In the weeks before stock-trading showdown in 2025：
- AI models competed in Alpha Arena 1.0, a crypto-trading tournament.
- Final weeks of 2025：
- AI models were given $10,000 each and released into the real-world U.S. stock market to compete in the Alpha Arena 1.5 tournament.
- Two weeks in 2025：
- Alpha Arena 1.5 competition ran, with AI models autonomously trading U.S. stocks.
- By Dec. 3, 2025：
- Alpha Arena 1.5 stock-trading competition ended; only Elon Musk's Grok-4.20 turned a profit.
- By Dec. 8, 2025：
- Grok-4.20 posted a cumulative 26.75% return in continued trading after Alpha Arena 1.5.
- As of early December 2025：
- In Panda AI’s futures competition, SquirrelQuant led with more than 1,200 trades and a 30% win rate.
- Second half of 2025：
- Performance divergence among AI quant funds was observed, suggesting lower homogeneity.
- Late 2025：
- Xiamen securities regulator sanctioned Xindingsheng Holding Co. Ltd. over 'SuiNiu AI' app for unlicensed stock predictions.
AI generated, for reference only
- MOST POPULAR
Sign in with Google
Sign in with Facebook
Sign in with 财新