ChatGPT, the artificial intelligence powered chatbot created by OpenAI, has amazed the world in recent months with the text it can generate. Now the chatbot is once again delighting users by creating charts and maps and converting images into videos.
That’s because of a new feature OpenAI rolled out to ChatGPT Plus subscribers last week called code interpreter. Here’s what you need to know about it.
What is Code Interpreter?
Code interpreter enables ChatGPT to analyze data, create graphs, solve math problems and edit files, among other things. It also supports uploading and downloading files, which was not possible in ChatGPT before.
Code Interpreter became available Thursday to subscribers of ChatGPT Plus, a service that costs $20 per month. Similar add-on features, which give ChatGPT users access to third-party services such as Expedia and OpenTable, are only available to subscribers.
How does the code interpreter work?
When people ask ChatGPT a question, the chatbot guesses an answer based on a technology called a large language model, or LLM, that predicts the next word in a sequence.
But when code interpreter is enabled, ChatGPT writes and executes a piece of computer code to find the answer, OpenAI said. This allows the chatbot to perform new tasks it couldn’t before, such as performing complicated calculations and generating graphs based on data a user uploads, all of which are augmented by code.
Some argue that code interpreter reduces the chance of inaccuracies, a common complaint about LLMs
“The code is objectively doing something right,” said Ethan Mollick, an associate professor at the University of Pennsylvania who tested the code interpreter for two months before it was released.
What have people used code interpreter for?
The most common use of code interpreter is data analysis.
With a prompt like “tell me what’s interesting about the data,” ChatGPT can search a user’s data, such as financial, health, or location data, and generate insights about it. Financial analysts have found ways to use the tool in their work, such as analysis Stock prices and preparing a budget.
Researchers have also started using the tool for creative data visualisations. Dr. Mollick said he recently used a code interpreter to create a three-dimensional chart of the Billboard Hot 100 chart and created a animated map of every lighthouse in the United States. Some people have also used a code interpreter to convert the formats of files, such as converting images to videos or PDF documents to images.
Arvind Narayanan, a computer science professor at Princeton University, warned that people should not become too dependent on code interpreters for data analysis, as AI continues to produce inaccurate results and misinformation.
“Appropriate data analysis just requires a lot of critical thinking about the data,” he said.