Nasdaq Gets SEC Nod for AI-Driven Trading Order

On September 8, Nasdaq declared that the United States Securities and Exchange Commission (SEC) had approved its request to introduce the first exchange order type run by artificial intelligence.

The new system, known as the dynamic midpoint extended life order (M-ELO), expands on the M-ELO automated order type by making it “dynamic,” i.e., using artificial intelligence to update and recalibrate itself in real-time.

Order types are a collection of software instructions that implement particular trade pairs at precise market price thresholds.

This form of automation has existed for some time, but the new AI-driven order type is the first of its kind to implement orders using AI with reinforcement learning in real time. This should result in a significant acceleration of orders submitted through the system.

Nasdaq stated in a blog post accompanying the approval announcement that dynamic M-ELO demonstrated a “20.3% increase in fill rates and an 11.1% reduction in mark-outs” during testing.

According to a Nasdaq data sheet, by adjusting the holding periods for orders in real-time, as opposed to the traditional system that merely applies static timeouts to orders, fill rates should increase without a corresponding increase in market impact. He said:

“Calculated on a symbol-by-symbol basis, this new functionality analyzes 140+ data points every 30 seconds to detect market conditions and optimize the holding period prior to which a trade is eligible to execute.”

The introduction of artificial intelligence technologies within the fintech industry has affected the entire financial sector.

ChatGPT and comparable large language models have been adapted for use as educational resources for both traditional stock traders and cryptocurrency traders.

Nasdaq’s previous attempts to combine AI with finance included the incorporation of predictive AI models to assist in parsing the more than 1.5 million options listings on the U.S. market.