Code Interpreter became available to all ChatGPT subscribers – “now everyone can become a data analyst”

OpenAI first announced third-party plugins for its popular ChatGPT service back in March, allowing users to extend its functionality by reading PDF files, for example. The company announced this week that it is releasing one of its own plugins – Code Interpreter – and making it available to all ChatGPT Plus subscribers.

The Code Interpreter “allows ChatGPT to run code, optionally having access to the files you upload,” an OpenAI spokesperson wrote in a blog post. “You can ask ChatGPT to analyze data, create graphs, edit files, perform mathematical calculations, etc.”

With a wide range of tools and a large amount of memory, AI can write code in Python and work with files up to 500 MB in size.

Code Interpreter allows ChatGPT Plus users to create charts, maps, data visualizations and graphs, analyze music playlists, create interactive HTML files, clean up datasets, and extract color palette from images. The interpreter opens up many possibilities, making it a powerful tool for visualizing, analyzing, and manipulating data.

Unsurprisingly, initial responses from experienced ChatGPT users and technology influencers have been overwhelmingly positive.

New features to use

Linas Beliunas, Europe and Lithuania Manager at Flutterwave, wrote a review on his LinkedIn: “OpenAI is bringing its most powerful feature since GPT-4 to everyone. Now anyone can become a data analyst.”

Belyunas attached a slideshow to his post showing 10 examples of new visualization and data analysis tasks he was able to solve with ChatGPT using the Code Interpreter, including creating an HTML interactive “heat map” of UFO sightings across the US using only ” raw data set”.

Ethan Mollik, Associate Professor at the Wharton School of the University of Pennsylvania and renowned AI expert, wrote in his Substack mailing list “One Useful Thing” that ChatGPT with Code Interpreter is “the most useful and interesting AI I’ve used.”

Mollick writes that the Code Interpreter “makes AI much more versatile” and can provide structured data that supports the user’s thesis: “For example, I asked him to prove to a doubter that the Earth is round using code, and he provided several arguments by combining text with code and pictures.”

The new use cases should also help OpenAI counter growing dissatisfaction from some users, in particular members of the ChatGPT and AI Reddit subreddits, who note that ChatGPT has become more limited and less functional over time, prohibiting certain topics and areas of conversation.

Safety first

Security remains a central element of Code Interpreter development. The main goal is to ensure that the code generated by AI does not lead to any unforeseen consequences in the real world. As users explore and discover new applications, OpenAI plans to continue to improve security protocols based on the knowledge gained during the beta.

Future of AI

One of the most intriguing applications of Code Interpreter is in data science, where it is said to operate at an “advanced level”. It can automate complex quantitative analyses, combine and clean data, and even reason about it in a human way. AI can create visualizations and dashboards that users can refine and customize by simply interacting with it. The ability to get downloadable results adds another layer of usability to Code Interpreter.

According to Mollik, this tool represents the most compelling evidence that AI is a valuable assistant in complex knowledge work. While human control remains essential, the new feature reduces the amount of routine work, providing more meaningful and in-depth work.

“Code Interpreter represents the clearest positive vision to date of what AI can mean for work. Yes, disruptive innovation, but innovation that leads to better, more meaningful work”, says Mollik.

Code Interpreter is clearly setting new standards for the future of AI and data science. With this tool, OpenAI is once again pushing the boundaries of ChatGPT and Large Language Models (LLMs) in general.”