Bhushan is a FinTech enthusiast and holds a good flair in understanding financial markets. His interest in economics and finance draw his attention towards the new emerging Blockchain Technology and Cryptocurrency markets. He is continuously in a learning process and keeps himself motivated by sharing his acquired knowledge. In free time he reads thriller fictions novels and sometimes explore his culinary skills.
The researchers also stated that ChatGPT could beat the commonly used models from Google, dubbed BERT along with the classifications based on dictionaries.
OpenAI’s popular chatbot system ChatGPT hasn’t failed to surprise people across the world, and once again, it’s out to show its AI prowess. Earlier this month, two new papers have been published that deployed the AI chatbot in some market-relevant tasks. The first task involved deciphering whether the Fed’s comments are dovish or hawkish. The other task determined for the ChatGPT experiment was whether the news headlines are good for the stock movement or not.
To anyone’s surprise, ChatGPT successfully aced both of these tests. Of course, this is nothing new for Wall Street which has used strong predictive models in place for a long time by using similar language models underpinning the chatbot. However, these findings convey that OpenAI has reached new levels of technological prowess in terms of parsing nuance and context.
In the first paper titled “Can ChatGPT Decipher Fedspeak?”, two researchers from the Fed confirmed that ChatGPt came the closest to humans in figuring out the statements from the central bank. Interestingly, the researchers also stated that ChatGPT could beat the commonly used models from Google, dubbed BERT along with the classifications based on dictionaries.
ChatGPT could also explain the classifications – whether dovish or hawkish – in a way that resembled the central bank’s own analyst.
ChatGPT Forecasting Stock Prices
The second study has the title “Can ChatGPT Forecast Stock Price Movements?” The researchers asked ChatGPT to pretend to be a financial expert and interpret the corporate news headlines. Thus, the study found that the answers given by ChatGPT included statistical links to the stock’s subsequent moves.
This was a clear-cut indication that the tech was successfully able to parse the news headlines. currently, the most-sophisticated quants in action use NLP to decide how popular stock is on Twitter, thus incorporating the latest headlines on a company.
But ChatGPT has demonstrated eye-popping advances by opening up a whole new world of information by making the tech accessible to millions of users across the globe.
Also, there’s no surprise that there are many machines that can read people, however, the thing with ChatGPT is that it can exponentially speed up the entire process. Slavi Marinov, the head of machine learning at Man AHL stated:
“It’s one of the rare cases where the hype is real. Previously (while using NLP), you had to label the data yourself. Now you could complement that by designing the right prompt for ChatGPT.”
Read other tech news on Coinspeaker.
Disclaimer: Coinspeaker is committed to providing unbiased and transparent reporting. This article aims to deliver accurate and timely information but should not be taken as financial or investment advice. Since market conditions can change rapidly, we encourage you to verify information on your own and consult with a professional before making any decisions based on this content.