The world of investing has seen major changes in recent years, driven by advances in technology. Automated trading and algorithmic investing have become increasingly popular, and the emergence of artificial intelligence has enabled investors to use powerful new tools to make smarter decisions. One of the most talked–about AI applications in the investing world is GPT–3, a natural language processing system developed by OpenAI.
GPT–3 is a powerful machine learning system that can generate human–like text based on inputs provided by the user. GPT–3 can be used to generate reports, analyze data, and make predictions about future market conditions. For investors, this means that they can use GPT–3 to automatically generate investment strategies and reports, freeing up their time to focus on other important tasks.
Experiment 1: Artificial Creativity
Because it’s the Christmas season, we’re in a festive mood. Since it’s been a tough year for the markets, we thought to bring some holiday cheer with an investing-themed Christmas carol. While songwriting may come naturally to some people, it might take us all day to write even two rhyming verses.
Now let’s see how AI performs:
Write a summary of the financial markets in 2022 in the style of a Christmas carol
Oh the markets of 2022, they are a sight to see
Stocks are up and the economy is healthy
The bull market is strong, to all a good night
As investors rejoice, with money so bright
GDP growth is soaring, rising to new heights
And the central bank’s stimulus is sure to delight
The Fed has done its job, a festive mood
As the market’s all-time highs, we can conclude
The yield curve is steep and the dollar is strong
Good news for investors who want to get along
The tech and biotech stocks, they are the best
The Nasdaq Composite and S&P 500, they pass the test
The global markets, they are strong too
The Euro and the Pound, they’re both in view
The Chinese economy has taken flight
A Christmas miracle, what a sight!
So raise your glasses and give a cheer
For the wonderful markets of 2022 this year.
At first glance this is great work! But it doesn’t take long to see the problem: 2022 wasn’t exactly a merry year for investing.
AI and ML techniques are highly deductive systems. Their output is only as good as the data used to train them. It appears that more of the text used to train this text generation mode was from a bull market. This might be expected since bull markets are more common and outlast bear markets. But AI often misses the mark when less data is available.
The S&P 500 index is down significantly year-to-date
Experiment 2: Blog Articles
Write an article about using gpt-3 for investing
We actually used the output from this prompt to write the entire introduction section of this article. This is unsurprising, because the content is generic and merely provides context about the technology we’re examining. The remainder of the article (shown below) is highly suspect. It sounds like articles written about just about any technology. One could easily replace “GPT-3” with “Computers” or “Trained Monkeys” and the article would still make sense semantically.
GPT-3 has been used by a number of organizations in the financial services industry, including hedge funds and asset managers. These companies use GPT-3 to generate automated investment strategies, analyze financial data, and create reports. The system can also be used to identify trends in the market and suggest potential investments.
Using GPT-3 for investing can help investors make smarter decisions and reduce the amount of time required to analyze data and generate reports. The system can also help investors stay up-to-date on the latest market trends and identify potential opportunities. Furthermore, GPT-3 can be used to generate investment advice tailored to an individual investor’s risk tolerance and financial goals.
Overall, GPT-3 has the potential to revolutionize the way investors make decisions. The system can help investors save time and make more informed decisions, enabling them to maximize their returns. With its ability to generate investment strategies and analyze data, GPT-3 is set to become an invaluable tool for investors in the years to come.
Conclusion: Not ready for prime time
To summarize, while GPT-3 is certainly impressive (especially with creative writing), it doesn’t appear like it can offer personalized financial advice anytime soon. AI and ML technologies are often considered to be highly applied statistical theory. Much of statistics revolves around analyzing the average within a dataset. GPT-3 might be able to offer average, generic advice about investing, but probably nothing more useful than common truisms.