AI Outperforms 93% of Fund Managers with 6X Higher Gains

Image Credit: Adam Śmigielski | Splash

Companies are increasingly prioritizing artificial intelligence investments in essential business areas, a recent IBM study shows, while independent academic research indicates AI systems can significantly enhance investment performance using publicly available information, highlighting the technology's evolving practical applications.

The IBM Institute for Business Value report, released in June 2025, found that 64% of AI budgets are now allocated to core functions such as supply chain management and human resources, compared to 36% for non-core activities. This reallocation aims to secure more sustainable returns following underwhelming results from initial generative AI efforts. Separately, a Stanford University study revealed an AI system that, over a 30-year simulation, outperformed 93% of human mutual fund managers by an average of 600%, generating six times the benchmark-adjusted returns.

Reallocation of AI Budgets

The IBM study, conducted in partnership with Oxford Economics, surveyed 2,900 executives across multiple industries and regions in early 2025. It identified a trend where organizations move away from peripheral projects toward integrating AI into fundamental operations. Core functions include areas like procurement and talent management, where AI can drive efficiency.

Initial pilots of generative AI in 2023 yielded a 31% return on investment, but scaled implementations have averaged only 7%, below the standard 10% cost of capital benchmark. In contrast, leading organizations report an 18% ROI by embedding AI strategically. Overall, just 25% of AI initiatives over the past three years have met expected returns, prompting this budgetary pivot.

Role of Agentic AI

Agentic AI, characterized as autonomous systems that manage intricate workflows with human collaboration, is gaining prominence in this shift. These systems enable independent task execution, differing from earlier automation by handling dynamic processes. The study notes that executives anticipate AI-enabled workflows to rise from 3% currently to 25% by year-end 2025.

Ad hoc AI pursuits have declined from 19% to 6% of organizations, reflecting a preference for coordinated strategies. Among firms adopting an "AI-first" mindset, 52% of revenue growth and 54% of margin improvements in the last year are attributed to AI.

AI in Investment Management

The Stanford research, led by professors Ed deHaan and Suzie Noh with collaborators Chanseok Lee and Miao Liu, examined AI's ability to optimize stock selections using public data. Published as a working paper in 2025, it simulated portfolio adjustments for about 3,300 U.S. equity mutual funds from 1990 to 2020.

Trained on 1980-1990 market data, the AI model—employing techniques to predict performance—analyzed 170 public variables, including interest rates and earnings call sentiments. It rebalanced holdings quarterly, replacing underperformers while preserving risk and diversification. Human managers produced $2.8 million in quarterly alpha; the AI achieved $17.1 million, a sixfold increase.

Historical Context and Drivers

The IBM findings extend from post-2023 assessments of generative AI, where high initial excitement met implementation challenges. The focus on agentic AI addresses needs for reliable outcomes in competitive environments, as fragmented approaches yield inferior results in growth and satisfaction metrics.

The Stanford work stems from inquiries into the untapped potential of public financial data, which humans struggle to process comprehensively due to volume and complexity. By quantifying this "shadow value", the study illustrates AI's capacity to bridge informational gaps without exclusive insights.

Potential Effects on Industry and Economy

Budget shifts to core operations may improve efficiency and profitability, with AI-attributed gains in operating profits observed since 2022. In finance, AI's demonstrated edge could lower costs in portfolio management, potentially reducing demand for entry-level analytical roles.

Broader consequences include integration hurdles for agentic systems, where coordination demands could slow adoption. In markets, if AI tools become ubiquitous, they might equalize advantages derived from public data, altering competitive dynamics.

Advantages and Drawbacks

Benefits encompass enhanced scalability: Agentic AI in operations supports enduring margin growth, while in investments, it delivers superior alpha at minimal additional cost through efficient data handling.

Challenges involve oversight risks, as autonomous errors in workflows could compound problems. In stock analysis, models may miss contextual nuances, introducing vulnerabilities, and workforce shifts could occur, though hybrid models might redirect human efforts to oversight.

Outlook for AI Developments

Executives project AI spending to grow from 12% to 20% of IT budgets by 2026. Agentic AI is expected to proliferate, reshaping processes across sectors, albeit with measured rollout to manage complexities.

In finance, automation trends suggest expanded use in data-intensive functions, possibly redefining fund management. Experts note that as adoption spreads, effectiveness could stabilize, underscoring the value of combined human-AI frameworks amid regulatory considerations.

License This Article

Source: IBM, Stanford Report

$5.00

$10.00

$20.00

$30.00

Custom Amount

Please enter an amount

$

One-Time Donation Monthly Donation Quarterly Donation Annual Donation

3% Cover the Fee

Support us to keep delivering insightful content on AI development. Your contribution fuels our mission to inspire and inform! Support us to keep delivering insightful content on AI development. Your contribution fuels our mission to inspire and inform!