OpenAI has made significant strides in the world of artificial intelligence with its latest offering, a feature known as the “code interpreter” for its ChatGPT Plus subscribers. This new capability marks a major development in AI technology, bringing an array of new possibilities to the fingertips of users.

The code interpreter is a tool that empowers the ChatGPT AI to perform several advanced functions. These include the capacity to analyze data, solve mathematical problems, generate charts, edit files, and even support the uploading and downloading of files. This marks a departure from the previously available capabilities of ChatGPT and introduces a higher level of versatility and utility for the users. This feature, which comes at a cost of $20 a month, effectively expands the utility of the AI, integrating it more seamlessly into various workflows.

Unlike the traditional functionality of language models, which predicts the next word in a sequence based on previous words, the code interpreter takes a slightly different approach. When a question is posed to ChatGPT with the code interpreter enabled, the AI writes and runs a piece of computer code to find the answer. In essence, it allows the AI to interpret and execute commands in a programming-like fashion to accomplish new tasks. This includes the performance of complex calculations, data analysis, and even the generation of charts and visualizations based on user-uploaded data.

This functionality has been said to reduce the chances of inaccuracies, a common complaint about large language models. Ethan Mollick, an Associate Professor at the University of Pennsylvania, emphasized the objectivity of the code interpreter. He tested the feature for two months before its release and expressed that “the code objectively does something right.”

The most common application of the code interpreter so far has been in data analysis. Upon instruction, ChatGPT can examine a user’s data, which can range from financial records to health statistics or geographic information. By doing so, the AI can generate relevant insights and predictions based on the data. The utility of this feature extends into professional domains as well. Financial analysts, for instance, have found use cases for the tool in analyzing stock prices and in budgeting.

Researchers have also harnessed the power of the code interpreter for creating innovative data visualizations. Dr. Mollick mentioned that he used the feature to create a three-dimensional chart of the Billboard Hot 100 list and make an animated map of every lighthouse in the United States. Other users have exploited the tool for file format conversion, such as transforming images into videos or converting PDF documents into images.

Despite these advancements, however, experts caution users not to become overly reliant on the code interpreter. Arvind Narayanan, a Professor of Computer Science at Princeton University, warns that while AI tools like the code interpreter can be useful, they are not infallible. There is still a risk of inaccurate results and misinformation, as AI still lacks the critical thinking ability that is central to appropriate data analysis. This reminder underscores the need for users to apply their judgment when using AI tools like the code interpreter.