Truace
Subscribe
55/100

Moderate evidence

Cite
More
Export CSVExport JSONEmailLinkedInX

Study finds AI hedge funds beat peers until 2017, then lost the edge

Study title: The Growth and Performance of Artificial Intelligence in Asset Management

Evidence type
Observational studyPreprint, not yet peer reviewed
Sample
n = 7,896
Sources
1 sourcePeer-reviewed
Published
3 min read
Truvace record
TRV-2026-2500cite or check it

Study title: The Growth and Performance of Artificial Intelligence in Asset Management

Economists at the University of Melbourne, the University of Texas at Austin and Southern Methodist University tracked 7,896 US hedge funds from 2006 to 2024. AI-driven funds beat non-AI funds by about 6% a year on a benchmark-adjusted basis in the early years. Before 2017 the gap was 49.6 basis points a month, and it disappeared afterward.

Read the study abstract

We examine the growth and performance of AI-driven investing. Using investment advisers' regulatory disclosures, labor market data, and fund strategy descriptions, we document that AI-driven investing has grown steadily since the early 2010s and is concentrated among hedge funds. AI hedge funds outperformed non-AI hedge funds in the early years, but this outperformance declined over time, even among early adopters. Contrary to concerns about AI-driven strategy homogeneity, AI hedge funds exhibit lower return com…

The Growth and Performance of Artificial Intelligence in Asset Management
"WCUGU Board of Directors' meeting 2025 02" by UgandaWiki, CC BY-SA 4.0.

50Gain side

How the two sides scoreClose call

51Problem side

The gain

In a study of 7,896 US hedge funds, funds that disclosed AI-driven strategies beat non-AI funds by about 6% a year after benchmark adjustment in the early years.

The problem

The same study found the AI funds' edge fell to statistically indistinguishable from zero after 2017, even for funds that adopted AI early.

A fixed formula sets both scores from the evidence and its sources. Reader votes don’t move them.

What the scores mean

G is the gain side, out of 100. It goes up when the gain is specific, when it measurably helped people and when the sources behind it are strong.

P is the problem side, out of 100. It goes up when the problem is specific, when it measurably hurt people and when the sources behind it are strong.

In brief

The fade was not explained by early versus late adopters: funds launched as AI funds in the early years lost most of their advantage after 2017. The authors link the decline to markets adapting as AI strategies spread and exploit the same opportunities.

Policymakers have worried that AI trading would make funds move in lockstep. Among funds in the same strategy category, AI funds' returns were less correlated with each other (0.021) than non-AI funds' were (0.124). The work is a working paper that has not been peer reviewed, and it measures only disclosed AI use.

Main points

  1. Only 89 of 7,896 hedge funds (1.1%) were ever identified as AI funds, a deliberately strict definition; AI funds' assets reached about USD 12 billion by 2024.
  2. AI funds outperformed non-AI funds by 49.6 basis points a month before 2017, and the difference disappeared after 2017.
  3. Compared with sibling funds run by the same adviser, AI funds outperformed by 34.9 to 41.1 basis points a month over the whole sample, so adviser quality alone does not explain the gap.
  4. Average pairwise alpha correlation was 0.021 among AI fund pairs versus 0.124 among non-AI pairs in the same strategy category.
  5. Investors did not send extra money to AI funds beyond what past performance explained; a one percentage point rise in lagged 12-month alpha went with 34.5 basis points higher monthly inflow.
  6. About 60% of AI funds in 2024 followed systematic diversified macro strategies, and AI-Autonomous funds held over 90% of AI fund assets.
  7. AI funds charged incentive fees nearly 2.5 percentage points higher but imposed roughly 60% shorter lockup and restriction periods.
  8. AI and non-AI funds generated similar net value added over their lifetimes, because AI funds' higher alpha was offset by smaller assets.

The full story

The authors built the sample from SEC Form ADV filings (116,276 brochures covering 62,253 adviser-years from 2012 to 2024), Lightcast job postings and archived Hedge Fund Research strategy descriptions. They used OpenAI's GPT-5 to classify filings and strategy text, applying a strict prompt to limit false positives. They defined AI-driven investing as autonomous AI strategies or strategies in which AI signals directly inform investment decisions.

Advisers mentioning AI in their filings grew from fewer than 10 in 2012 to nearly 300 in 2024, but most of the 2023 to 2024 jump came from risk disclosures, AI-themed assets and research support. Among about 1,550 advisers disclosing quantitative strategies in 2024, around 100 described AI-autonomous or AI-augmented strategies. AI-related job postings stayed below 4% of postings and rose sharply around 2017, with advisers showing higher AI labor intensity than other firms.

Advisers that manage hedge funds, and advisers with larger private-fund assets, posted more AI jobs. AI funds clustered in macro strategies, where liquid markets suit short-term trading despite small datasets. They showed lower exposure to equity risk factors and a positive loading on momentum.

On performance, AI funds' outperformance fell by 5.9 basis points a year on average. Using AQR factors gave similar but slightly smaller results, and the early outperformance was concentrated among AI-Autonomous funds. Early-adopter AI funds beat non-AI funds before 2017, but their edge vanished on one benchmark and reversed on the other afterward, while late adopters showed no significant difference. The authors say the findings point to alpha decay and to growth in AI and quantitative investing making persistent outperformance harder to sustain.

Sources

1 on file
  1. NBER Working Paperdoi:10.3386/w35273
    Peer-reviewed2026-06-01

Talk it over

Four questions to start onThis record found a gain and a problem in the same place. Which matters more to you, and who might answer differently?

Start with a question

Study finds AI hedge funds beat peers until 2017, then lost the edgehttps://truvace.com/item/the-growth-and-performance-of-artificial-intelligence-in-asset-management

  1. This record found a gain and a problem in the same place. Which matters more to you, and who might answer differently?

    The gain In a study of 7,896 US hedge funds, funds that disclosed AI-driven strategies beat non-AI funds by about 6% a year after benchmark adjustment in the early years.

    The problem The same study found the AI funds' edge fell to statistically indistinguishable from zero after 2017, even for funds that adopted AI early.

    Answer in the thread

  2. How do we know? What can this kind of evidence show, and what can’t it?

    The evidence This rests on an observational study with a sample of 7,896, from NBER Working Paper. It is a preprint, not yet peer reviewed.

    Answer in the thread

  3. What does it leave open? What else would you want to know before acting on it?

    Answer in the thread

  4. Have you seen AI used this way where you study, work or live? Did it go the same way?

    Answer in the thread

The questions are the same on every record. The details under them are taken from this record as written.

What readers say

More from Business