Do AI and Machine Learning Funds Actually Outperform?
Evaluating AI & ML-focused U.S. Equity Funds
September 2026. Reading Time: 10 Minutes. Author: Nicolas Rabener.
- AI & ML-focused funds performed poorly
- AI-focused hedge funds have also underperformed
- This explains the few funds and low AUM
INTRODUCTION
In October 2019, we conducted our first review of funds using AI for investment decisions, where the performance was rather unimpressive (read AI, What Have You Done for Me Lately?). In February 2022 and July 2023, we reviewed these funds again, along with new entrants, but the picture hadn’t changed (read Smart Money, Crowd Intelligence, and AI and Hitting Home Runs with AI Investing?). The universe included ETFs and hedge funds using artificial intelligence for stock selection and asset allocation decisions.
Fast forward to today, and we’re conducting another review. AI has permeated daily life and empowered fund managers to analyze companies and markets in seconds rather than days or weeks. Theoretically, funds harnessing these new powers should be entering a golden age of consistent alpha generation.
Let’s review whether theory and reality match up.
FUNDS OVERVIEW
In this analysis, we review U.S. equity funds using AI and machine learning (ML), as well as AI-driven hedge funds. For the avoidance of doubt, these funds do not provide exposure to AI companies; they use AI and ML for investment decisions.
The first surprising finding is that the number of clearly identifiable AI and ML funds is rather low. There are 9 ETFs focused on using AI, but two of these have already been liquidated. Some funds, like the WisdomTree International AI Enhanced Value Fund (AIVI) and WisdomTree U.S. AI Enhanced Value Fund (AIVL), were repurposed to AI-driven strategies from different original mandates.
It is hard to identify funds that explicitly use ML. BlackRock is one company that has been vocal about deploying ML within its Systematic Active Equity group across a handful of funds since 2014. Voya Investment Management has its own Voya Machine Intelligence (VMI) team, but it offers strategies only via managed accounts, where no public return data is available. One ETF, the Simplify Wolfe US Equity 150/50 ETF (WUSA), used ML explicitly, but was liquidated after seven months – hardly a sign of commercial success.
The universe comprised only 14 funds, managing a total of $12.7bn, but almost all of this belongs to the BlackRock funds. The total AUM of pure AI-focused funds is less than $1bn – tiny compared to the capital flowing into AI model makers, data centers, semiconductors, and related infrastructure.
PERFORMANCE OF AI & ML FUNDS
The low asset levels in AI- and ML-focused funds suggest poor performance. We compute excess returns by subtracting each fund’s benchmark return, then show the excess returns of the worst, average, and best funds. We observe that both AI and ML funds have underperformed their benchmarks on average.
The case could be made that AI was far less developed before ChatGPT launched in 2022, but these funds have not improved since then. For example, the Amplify AI Powered Equity ETF (AIEQ) has been using IBM’s AI, namely Watson, to select a diversified portfolio of U.S. stocks across market capitalizations. Yet the fund’s underperformance has increased even as AI models have improved dramatically.
PERFORMANCE OF AI HEDGE FUNDS
Perhaps ETFs and mutual funds are too constrained to fully exploit AI and ML. We can test this thesis by comparing the performance of the Eurekahedge AI Hedge Fund Index against the broader With Intelligence Hedge Fund Index. The Eurekahedge AI Hedge Fund Index is being decommissioned, with the last available return data from December 2025. This is not a positive sign and likely reflects AI hedge funds significantly underperforming the broader hedge fund universe.
Source: Finominal
FURTHER THOUGHTS
It is curious that the rapid evolution of AI models has not enabled AI and ML-focused funds to generate significant outperformance. The irony is that these funds would only have needed to bet on themselves – via companies like NVIDIA – and their performance would have been far better.
RELATED RESEARCH
Hitting Home Runs with AI Investing?
Smart Money, Crowd Intelligence, and AI
AI, What Have You Done for Me Lately?
Less Efficient Markets = Higher Alpha?
Alpha Generation: Equity Generalists vs Sector Specialists
EM Small-Cap Funds: A Niche Alpha Story?
EM Hedge Funds: Extracting Alpha from Inefficient Markets?
Betting on Insiders
ABOUT THE AUTHOR
Nicolas Rabener is the CEO & Founder of Finominal, which empowers professional investors with data, technology, and research insights to improve their investment outcomes. Previously he created Jackdaw Capital, an award-winning quantitative hedge fund. Before that Nicolas worked at GIC and Citigroup in London and New York. Nicolas holds a Master of Finance from HHL Leipzig Graduate School of Management, is a CAIA charter holder, and enjoys endurance sports (Ironman & 100km Ultramarathon).
Connect with me on LinkedIn or X.
