ChatGPT has in recent months taken the world by storm with it’s AI powered language comprehension capabilities. The technology developed by OpenAI uses machine learning algorithms to analyse and understand natural language to generate human-like responses to a wide range of prompts and questions.
The technology based on GPT (Generative Pre-trained Transformer) architecture, pre-trained on large amounts of text data to learn patterns and relationships in language. Sounds like Natural Language Processing (NLP) no?
Here at VoxSmart we decided to put it to the test to see how it would fair with jargon heavy trading language used across platforms globally! We sat down with ex-broker and current Senior Business Development and Sales Representative Andrew Stone to coach us through trade talk! Let’s see how it did…
First in the fire…
Not the most promising start from the technology – So what does it mean Andrew?
From my understanding it refers to an exchange traded asset and the action of selling right at the end of the trading session to move the price down. This is a common occurrence on any trading floor and a phrase like this or similar can be heard daily as traders seek to strike the best deal.
Okay let’s try this, as suggested by the Essex native and cockney rhyming slang extraordinaire.
Is that how you used it?
No not quite – in my trading days in London a “bag” referred to a grand or 1000! There was a lot of folk from London on the desk, so using slang like this was part of our day-to-day lexicon. Those not familiar with rhyming slang had to quickly learn the jargon to keep up!
Okay, so let’s see if ChatGPT understands the term in relation to rhyming slang…
Nope, not even close! Or well I certainly have never of heard of it used in that way – a cup of Rosie Lee or simply a cup of Rosy perhaps, but cup of bag? Never!
Okay so it needs some training on certain cockney rhymes that could be used in trading, how does is deal with jargon used for information requests?…
So, what might this mean?
Typically, if someone were to ask you any cares on something it means they are looking for information on the supply and demand of a certain stock. For example, if someone were to ask “any cares in the Aug 25 Bund?” it means they are looking for a decent clip, actually – check if it understands what a clip is!
How did it do?
Yes, all relevant answers, which goes to show that although it may not be 100% attuned to jargon used on trading floor it still uses powerful technology to filter and find relevant information. However, current technology such as NLP which we use across our products is pretrained to understand trading specific jargon used across asset classes and capital markets.
There you have it! When it comes to trading specific jargon and slang ChatGPT has a little way to go to catch-up. Technology such as NLP as used across our product suite has been trained to understand the everyday lexicon of traders to ensure that business can cut through the noise of communications and make better business decisions!
Get in touch with us today to find out more on how we harness the power of AI today.
If you liked this, then you might also be interested in: