Can AI help investors achieve their wealth goals?

The fact that the technology can be used autonomously by investors without guidance generates a number of risks

A generative AI platform will return general results to queries about the investment process, which might not be in line with an investor’s real needs. Reuters

Generative artificial intelligence constitutes a revolutionary advancement in how data is collected and interpreted, being able to generate personalised, original output in relation to the request that is directed to it.

The disruptive impact such technology will have on the financial sector is evident, as the behaviour of investors will change as a result of the ever-increasing use of the technology and all the tools that are being built — and will continue to be built — on it.

The advantages are evident. By being able to not only gain access to virtually all the information available online at lightning speed, but to also have it structured in a format that adheres to the particular request of any single user, investors can gather the information they need to make investment decisions.

They can even obtain a complete investment plan, which might include an indication of the assets to be included and their weighting in the portfolio.

Furthermore, the investor will have access to all of the above for a fraction of the cost linked to professional analysis and advice, dramatically widening the reach to those whose wealth did not justify bearing the costs of professional advice.

Even more wealthy and sophisticated investors might migrate to new tools based on generative AI.

Forecasts have already been made with regards to the impact it will have on employment in the financial services space. And they are dramatic.

This is interesting, but the fact that generative AI can be autonomously used by investors without guidance generates a number of risks that we must be aware of to use it appropriately.

First of all, to properly formulate our queries to the system, obtain relevant and reliable results and, ultimately, be able to interpret the results, we need to know what we are talking about in the first place.

The relevance, reliability and effectiveness of generative AI is directly correlated with the level of accuracy and detail of the request that is fed to the system.

General queries about the investment process or about a certain part of it will return very general results, which might not be in line with the investor’s real needs.

To correctly execute the above, besides understanding the dynamics and components of the investment process, we must be able to identify and quantify the basic elements that allow a proper financial planning of our own assets.

We need to evaluate our actual and future financial flows, identify our investment goals and risk tolerance.

Delegating such assessment to a third party, generative AI in this case, might be potentially harmful and lead to serious disappointment in the investment process outcomes.

The system itself warns about such risks and advises users to seek professional advice.

Professional advice is, therefore, not pushed (at least not completely) out of the door.

For all the instances involving insufficient knowledge of financial matters and terminology, time constraints or complexities in the level of the wealth to be analysed (which could bring in tax implications, overlapping with professional or corporate involvement, particular needs of wealth protection, et cetera), the guidance of an expert will still be needed.

This is where one can offset the impact of AI on financial services employment — by embracing a holistic approach to wealth management, refraining from resisting this new technology and, instead, embracing and harnessing its power.

We also must have awareness of unforeseen events that might radically change the scenario that we used to build our investment portfolio.

Black swan events are unpredictable, both in terms of timing and magnitude of their impact.

The very characteristics of such events do not allow us to identify a relevant and homogenous set of data on which to perform statistical analysis, recognise patterns, learn from the past and meaningfully reduce the risk of unpredictable and disappointing outcomes for our investment portfolio.

It is evident that generative AI can be extremely helpful in researching the single components of an investment portfolio, calculating the optimal weight within the portfolio, as well as analysing returns, volatility and possible outcomes based on different scenarios.

These are all features that were reserved for the most advanced research teams within investment banks and hedge funds but will be soon at the disposal of every one of us.

Exciting indeed, but we always need to be aware of the process and avoid excessively relying on technology. This is a basic risk management rule that is disregarded way too often.

Roberto d’Ambrosio is the chief executive of Axiory Global