The Limitations of Using AI as a “Free” Portfolio Tool

Generative AI platforms such as ChatGPT, Anthropic, and others have quickly become popular tools for brainstorming ideas, summarizing information, and assisting with research. Naturally, some financial professionals have begun exploring whether these tools can also be used for portfolio construction.

On the surface, the idea seems appealing. If AI can generate investment ideas and process large amounts of information, why not simply use it to build and optimize portfolios?

Well, in our opinion, there are several practical limitations that make general-purpose AI tools difficult to rely on as a solution for portfolio construction. While these tools can certainly complement the investment process, they were not designed to function as a portfolio management engine.

Let’s take a closer look at where some of the key challenges emerge.

Reliability & Mathematical Consistency

Generative AI models are designed to produce probabilistic language responses, not deterministic financial calculations. This distinction is important. Because these models generate responses based on probabilities, the same prompt can often produce slightly different outputs each time it is run. While this may be perfectly acceptable for brainstorming or drafting written content, it can present challenges when dealing with portfolio mathematics.

For advisors managing client assets, consistency and repeatability are essential. If a portfolio optimization is run today and again tomorrow under the same assumptions, the result should be identical. In our opinion, this level of repeatability is difficult to guarantee when relying on a general-purpose AI model.

We have typically noticed these variations are rarely trivial. This can have meaningful impact on the results you’re receiving, and therefore, the portfolios that you ultimately implement.

Portfolio Optimization Requires Specialized Infrastructure

Another important consideration is that generative AI models do not inherently contain the underlying infrastructure required to construct optimized portfolios.

For example, most portfolio construction frameworks rely on:

• risk modeling

• constraint-based optimization

• return assumptions

• portfolio mathematics grounded in modern portfolio theory

While AI can certainly help interpret instructions or organize information, it does not inherently provide the optimization engine required to translate investment views into actual portfolio allocations. In practice, building portfolios still requires a system specifically designed to run those types of calculations.

Data Integrity & Market Inputs

Another challenge is the sourcing and reliability of financial data. Generative AI platforms generally do not provide direct access to reliable, real-time market data. As a result, advisors attempting to construct portfolios using these tools often need to:

• manually source market data

• format the data for input

• validate the outputs independently

In our opinion, this introduces additional operational risk and can quickly become time consuming. Ideally, a portfolio construction tool should integrate the necessary market data and analytics directly into the workflow so that advisors can focus on interpreting results rather than assembling the inputs. In fact, I had experience attempting portfolio research with ChatGPT, only to realize that the data it was basing its thoughts on was from 2024.

Compliance & Fiduciary Considerations

Of course, advisors operate in a regulated environment, and that brings another important layer to this discussion. Portfolio decisions must be defensible, documented, and repeatable. Regulators are increasingly paying attention to how firms use artificial intelligence in their investment process, and the SEC has already begun emphasizing the need for formal AI governance policies.

General AI tools can make it difficult to demonstrate:

• how a portfolio recommendation was generated

• whether results are repeatable

• whether a consistent investment process is being followed

Workflow Efficiency

While generative AI tools may appear “free,” the true cost often emerges in the time required to make them usable within an investment workflow. Therefore, advisors may find themselves spending significant time:

• structuring prompts

• preparing datasets

• validating outputs

• recreating analyses for different scenarios

In contrast, purpose-built portfolio platforms can automate many of these steps. This allows advisors to spend less time constructing portfolios and more time focusing on the interpretation of results and client communication. It is also important to consider how these tools integrate with the rest of an advisor’s technology stack. Seamless data sharing between portfolio tools, custodians, and other parts of an advisor’s stack is critical for both efficiency and operational risk management.

Long-Term Stability

Finally, generative AI platforms are evolving extremely quickly. Capabilities, pricing models, and API access can change frequently. While this pace of innovation is exciting, it can also introduce uncertainty if a firm relies on these tools as a core component of its investment infrastructure.

In our opinion, systems designed specifically for portfolio management tend to provide greater stability and reliability because they are built around long-term advisor workflows rather than general-purpose experimentation.

The Bottom Line

Generative AI can be a valuable tool for research, idea generation, and improving workflow efficiency. However, when it comes to portfolio construction and optimization, advisors typically benefit from tools that were designed specifically for that purpose. Purpose-built portfolio platforms combine optimization engines, financial data, and integrated workflows to translate investment views into portfolio allocations in a consistent and repeatable manner.

While AI will almost certainly play an increasing role in investment management, in our opinion it works best as a complement to a dedicated portfolio system rather than as a replacement for one.

Disclosure

The reader should not assume that investment decisions identified and discussed were or will be profitable. Specific investment advice references provided herein are for illustrative purposes only and are not necessarily representative of investments that will be made in the future.

*This document does not constitute advice or a recommendation or offer to sell or a solicitation to deal in any security or financial product. It is provided for information purposes only and on the understanding that the recipient has sufficient knowledge and experience to be able to understand and make their own evaluation of the proposals and services described herein, any risks associated therewith and any related legal, tax, accounting or other material considerations. *

Certain information contained herein has been obtained from third party sources and such information has not been independently verified by Global Beta Advisors (“Global Beta”). No representation, warranty, or undertaking, expressed or implied, is given to the accuracy or completeness of such information by Global Beta or any other person. While such sources are believed to be reliable, Global Beta does not assume any responsibility for the accuracy or completeness of such information. Global Beta does not undertake any obligation to update the information contained herein as of any future date.

Except where otherwise indicated, the information contained in this presentation is based on matters as they exist as of the date of preparation of such material and not as of the date of distribution or any future date. Recipients should not rely on this material in making any future investment decision.

Certain information contained herein constitutes “forward-looking statements,” which can be identified by the use of forward-looking terminology such as “may,” “will,” “should,” “expect,” “anticipate,” “project,” “estimate,” “intend,” “continue,” or “believe,” or the negatives thereof or other variations thereon or comparable terminology. Due to various risks and uncertainties, actual events, results or actual performance may differ materially from those reflected or contemplated in such forward-looking statements. Nothing contained herein may be relied upon as a guarantee, promise, assurance or a representation as to the future.

The results discussed herein are derived from both quantitative and qualitative factors, including historical returns and market conditions and assumptions. The projected results are presented to establish a benchmark for future evaluation of its performance, to provide a measure to assist in assessing the anticipated risk and reward characteristics of an investment to facilitate comparisons with other investments. Any target data or other forecasts contained herein are based upon highly subjective estimates and assumptions about circumstances and events that may not yet have taken place and may never do so. If any of the assumptions used do not prove to be true, results may vary substantially. The projected investment returns are pre-tax and represent possible returns that may be achieved. The projected investment returns are subject to change at any time and are current as of the date hereof only. In any given year, there may be significant variation from these projections, there is no guarantee that they will be able to achieve the projected investment returns in the short term or the long term.

The results contained herein are for illustrative purposes only, do not represent the performance of any Global Beta Advisor Artha product or any particular investment, and are not intended to predict or depict future results. Performance does not reflect the deduction of fees or expenses, returns received by an investor would otherwise be lower.