Amazon unveils CodeWhisperer, a code-writing AI like GitHub’s Copilot

24 Jun 2022

Image: © MIND AND I/Stock.adobe.com

Amazon said CodeWhisperer uses ‘contextual clues’ to give recommendations that can save time and effort for developers.

Amazon has launched a preview version of an AI-powered programming tool designed to help developers write code, similar to GitHub’s Copilot.

CodeWhisperer is described as a coding companion that continually examines a developer’s code and comments, presenting “syntactically correct recommendations” based on the user’s coding style and variable names.

Trained on billions of lines of code and powered by machine learning, Amazon said the software can help save time and effort for developers and reduce repetitive work.

An AI-powered coding assistant that acts like predictive text for programmers has already been seen with GitHub’s Copilot, which was made publicly available to developers earlier this week. It has been used by more than 1.2m developers in technical preview.

GitHub said the proportion of total code being written by the AI is on the rise. Last October, the Microsoft-owned company revealed that Copilot was helping write up to 30pc of code on the platform and now it has said its AI assistant is helping write nearly 40pc of code.

Amazon said its CodeWhisperer machine learning software was trained on billions of lines of code from the company’s internal repositories, along with open-source repositories, API documentation and public forums.

Amazon Web Services (AWS) VP and chief evangelist Jeff Barr said the software uses “multiple contextual clues” to push its recommendations, including the cursor location, the code that precedes the cursor and code in other files of the same project.

A preview is available as part of the AWS IDE Toolkit and works with Python, Java and JavaScript. Amazon said CodeWhisperer supports multiple IDEs including VS Code, IntelliJ IDEA, PyCharm, WebStorm and AWS Cloud9.

Amazon unveiled CodeWhisperer this week at its Re:MARS AI conference in Las Vegas. This conference is also where the tech giant revealed it is working on technology that will give Alexa the ability to mimic virtually anyone’s voice, including a deceased relative.

The technology has raised eyebrows for its potential uses beyond just mimicking the voice of a dead person. For example, it could be used to deepfake the voice of living people for malicious purposes.

10 things you need to know direct to your inbox every weekday. Sign up for the Daily Brief, Silicon Republic’s digest of essential sci-tech news.

Leigh Mc Gowran is a journalist with Silicon Republic

editorial@siliconrepublic.com