Leading the Alternative World Order

Reshaping Perspectives and Catalyzing Diplomatic Evolution

Thursday, May 9, 2024
-Advertisement-
Foreign AffairsDo “artificial intelligence” tools threaten the work of financial analysts?

Do “artificial intelligence” tools threaten the work of financial analysts?

– Published on:

These analyzes paint an explanatory picture of the economic and financial scene in detail, helping many to build their investment decisions at any time based on the messages carried by these data and between the lines of official speeches.

But this old habit, which is one of the main traditional foundations for customers, “may change soon”, and the reason for that is generative AI tools like ChatGPT.. so how?

Economists at the Federal Reserve Bank of Richmond asked ChatGPT to analyze Fed data and concluded that the tool “shows strong Fedspeak rating performance” and even “outperforms other popular rating methods for models.” GPT,” they report in a recent post.

Thus, “bots can decode the mind of Federal Reserve Chairman Jerome Powell better than other systems available.”

And on whether or not this is seen as a positive thing, Financial Times editorial board editor Gillian Romaine Tate writes in an article that for hedge funds looking for a competitive advantage as well as financial managers who want to downsize their staff, this can be a good thing.

Although the aforementioned research paper highlighted the current difficulties in relying on these tools completely “without human supervision”, Jillian believes that in the future, and with the development and improvement of ChatGPT, all of this can change, so it’s no surprise that this new research also warns that the jobs of some economists may soon be at risk.

Threat to the jobs of economists

In statements exclusive to Sky News Arabia Economy, Nicolas Gradojevet, a professor in the Department of Economics and Finance at Canada’s University of Guelph, whose research interests include (artificial intelligence and big data analytics) , says:

It is possible that artificial intelligence systems will completely overtake some jobs in finance and economics in the near future. Jobs involving big data processing and analysis are already mostly driven by AI.

“It’s high-speed computerized trading (or high-frequency trading) in stock markets,” he adds, pointing out that “up to 80% of total stock market trading volume in the United States is generated by algorithmic trading”. the systems…and therefore the trade Traditional human selection of stocks accounts for somewhere between 10 and 20% of stock market activity.

“I believe that some research functions in finance and economics will also be overtaken by artificial intelligence, which will provide automated trading advice and recommendations,” continues the professor in the Department of Economics and Finance at the University of Guelph.

The referenced article, however, states that “ChatGPT should currently only be used under human supervision, given that while it can correctly answer 87% of the questions on a standardized test of economic knowledge”, it “is only not flawless and may still misclassify sentences or fail to grasp nuances that a human rater with experience in the field would.

The researchers think ChatGPT could have potential as a “helper,” but they also point out that reliance on AI can sometimes lead to failure, in part because its dataset is limited and unbalanced.

Building investment decisions

And as to whether investors can, now or in the future, rely on tools like ChatGPT to make their investment decisions, without the need for human supervision, Gradojevit points out:

Artificial intelligence systems such as ChatGPT are developing and are still in their infancy. These systems learn and adapt to new information, but they can still make serious mistakes. Personally, I wouldn’t rely on ChatGPT’s investment advice just yet.

“I tested this on very simple financial reporting tasks and for example produced fake references that weren’t there…Also keep in mind that ChatGPT requires some intervention and constant human supervision, and that humans can make mistakes,” he explains.

He explains that it will be some time before we can fully rely on ChatGPT guidance for investment decisions without human supervision, pointing out that “human supervision is still a big part of the AI ​​learning process. “.

Three risks

On the other hand, the editor of the editorial board of the Financial Times, during her aforementioned article, quoted another article on artificial intelligence, published in 2020 by Lilly Bailey and Gary Gensler, which claimed that if the Generative AI can provide incredible benefits in this context, however, it creates three significant risks to stability.

The first of these is the “obfuscation” that AI tools are opaque to all but their creators. It might be possible to remedy this by asking them to publish their internal guidelines in a standardized way, but this seems unlikely.

The second such risk is associated with “concentration” in the sense that generative AI is likely to be dominated by only two players with a competitor or two in China, and many services will then be built on this basis. ‘IA (..) if a bug appears in this rule, it can poison the whole system. In addition to the third risk related to “regulatory gaps” stemming from the fact that financial regulators seem ill-equipped to understand artificial intelligence.

The economy of the future

Based on the above, Nicolas Gradojevit, Professor in the Department of Economics and Finance at Canada’s University of Guelph, continues his conversation with “Economy Sky News Arabia” discussing the impact of these evolving technologies on the economic system in the future, given that:

The economy of the future will be partly managed by artificial intelligence. AI will play an important role in big data analysis, stock trading, bookkeeping, accounting (daily detailed record of all financial transactions), banking, customer service, research, etc. However, issues such as regulatory compliance, ethics, data security, AI interpretability, and fairness require human oversight.

“Humans will need to develop comprehensive governance systems for AI, including policy management, compliance monitoring, risk management, reporting and auditing. These policies will be needed to ensure that AI operates within its ethical, legal and performance limits.”

Read the Latest World News Today on The Eastern Herald.


For the latest updates and news follow The Eastern Herald on Google NewsInstagramFacebook, and Twitter. To show your support for The Eastern Herald click here.

Arab Desk
Arab Desk
The Eastern Herald’s Arab Desk validates the stories published under this byline. That includes editorials, news stories, letters to the editor, and multimedia features on easternherald.com.

Public Reaction

Subscribe to our Newsletter

- Gain full access to our premium content

- Never miss a story with active notifications

- Exclusive stories right into your inbox

-Advertisement-

Latest News

-Advertisement-

Discover more from The Eastern Herald

Subscribe now to keep reading and get access to the full archive.

Continue reading

Discover more from The Eastern Herald

Subscribe now to keep reading and get access to the full archive.

Continue reading