Wall Street's Data-Obsessed Investors Slow to Adopt Generative AI

The world's most sophisticated investors are known for their relentless pursuit of any technological edge. Yet when it comes to generative AI, the data-obsessed quant community is moving with uncharacteristic caution.

A comprehensive new survey reveals that a majority of systematic trading strategists at top-tier asset managers have not yet begun their generative AI journey. This hesitation from the industry's most analytical minds offers a telling snapshot of the technology's current limitations in high-stakes finance.

The findings suggest that for all the hype surrounding artificial intelligence, its practical application in complex investment strategies remains a work in progress. The data points to a deliberate, measured approach rather than a wholesale rush to adopt.

The data giant conducted an extensive survey of 151 quantitative investors between April and November of last year. The goal was to determine how these professionals have integrated generative AI tools into their investment research processes.

The results were clear: 54% of respondents do not use generative AI for investing. This finding aligns with a broader sentiment observed at industry conferences, where skepticism about the technology's market-beating capabilities has been prevalent.

While the quant world has embraced machine-learning techniques for years, the survey indicates that generative AI has not yet broken through. The hesitation is not due to a lack of awareness, but rather a strategic decision based on current capabilities.

• 54% of quant investors do not use generative AI
• Survey conducted from April to November last year
• 151 quant investors interviewed
• Focus on investment research integration

The primary barrier to adoption is not a lack of interest, but a fundamental issue of data structure and formatting. According to industry analysis, the slow adoption will be linked to data availability going forward.

Angana Jacob, global head of research data, explains that quant investors require their data to be cleaned and structured in a very specific way. This is due to the complex systems their strategies run on and the enormous amount of capital at stake if an error occurs.

"They're working in a very controlled research environment, models need to be explainable, models need to be repeatable."

The work required to prepare datasets for AI use is described as unglamorous but foundational. This meticulous preparation is essential for maintaining the integrity of strategies that manage billions of dollars in assets.

Beyond data challenges, there is a deep-seated skepticism about generative AI's ability to generate alpha—the excess return that outperforms the market. At a London-based conference, quant investors expressed doubts about the technology's value-add to their processes.

This sentiment was echoed by a UBS executive, who stated that AI is not going to help win the 'alpha war.' The consensus among many quant executives is that AI is not yet capable of consistently beating the market.

The lack of widespread use in the investing process is viewed by some as a sign of the diligence of these players. Rather than chasing trends, these investors are waiting for the technology to mature and prove its efficacy in real-world, high-stakes scenarios.

• AI not seen as capable of beating the market
• Skepticism about adding value to investing processes
• Focus on explainable and repeatable models
• Caution viewed as a sign of professional diligence

Despite the current slow adoption, there is underlying enthusiasm for what generative AI could achieve once the data infrastructure catches up. Companies are actively working to bridge this gap.

Angana Jacob's team is creating data products specifically designed for quants, which could increase AI adoption in the future. The goal is to make the technology more accessible and reliable for controlled research environments.

The industry is not alone in identifying this issue. Startups like Carbon Arc, founded by former Point72 data executive Kirk McKeown, are also focused on structuring datasets for easier ingestion into artificial intelligence models. This ecosystem of data providers is crucial for the next phase of adoption.

"It's a good thing, it shows their caution."

The current pause in adoption is not a rejection of the technology, but a strategic wait for the right tools and data to make generative AI a reliable partner in the quest for market outperformance.

The relationship between Wall Street's quants and generative AI is defined by caution, not rejection. The technology is viewed as promising but not yet ready for prime time in high-stakes investment strategies.

The primary hurdles are practical: data must be meticulously cleaned and structured to meet the rigorous standards of quantitative finance. This foundational work is seen as essential before any meaningful adoption can occur.

Ultimately, the industry's measured approach reflects its core philosophy. In a field where precision is paramount, the most data-obsessed investors are waiting for the tools to mature before they fully commit their capital and strategies to the new technology.