Microsoft GitHub launches AI-powered Copilot X, adopts GPT-4 model

Copilot X allows software developers to automatically complete comments and code, make personalisation

GitHub, Microsoft’s software development collaboration platform, on Wednesday announced an update that will use the power of GPT-4, an artificial intelligence (AI) model.

GitHub said Copilot X allows software developers to automatically complete comments and code, offering personalisation for every programming team, project, and repository.

The tool will include a chat interface to the code editor that is focused on developer scenarios. It “natively” integrates with Visual Studio Code, the source-code editor made by Microsoft and the development environment Visual Studio. With the new update, a developer can get in-depth analysis and explanations of what code blocks are intended to do, generate unit tests, and even get proposed fixes to bugs.

The open-source community is starting with documentation for React, Azure Docs, and MDN, to learn and iterate quickly with the developers and users of these projects. The new chat interface termed ‘docs’ will provide AI-generated responses to questions about documentation as well as the languages, frameworks, and technologies being used by developers.

The copilot chat will recognise what code a developer has typed, and what error messages are shown, and it is embedded into the integrated development environment (IDE) for storing, tracking, and collaborating on software projects.

The chat is built upon the work that OpenAI and Microsoft have done with ChatGPT and the new Bing. It will also join GitHub’s voice-to-code AI technology extension the company previously demoed, which is now called ‘Copilot voice’, where developers can verbally give natural language prompts.  

Developers can now sign up for a technical preview of the first AI-generated descriptions for pull requests on GitHub. Pull requests essentially refer to changes being made to a branch in a repository of GitHub, which is a developers’ community with millions of open-source projects.

GitHub Copilot is already writing on average 46 per cent of code in all programming languages, helping developers code up to 55 per cent faster within just two years of its launch.

The company is testing new capabilities internally, where Copilot will automatically suggest sentences and paragraphs as developers create pull requests by dynamically pulling in information about code changes. It is also preparing a new feature which enables Copilot to automatically warn developers if they miss sufficient testing for a pull request and then suggest potential tests that can be edited, accepted, or rejected based on a project’s needs.