GitHub Copilot, the controversial AI-driven coding tool will be generally available for developers later this summer, Microsoft announced on May 24. The tool will be initially available for free of cost only for students and verified open-source contributors.
The Copilot experience won’t change much with general availability. As earlier, developers will be able to cycle through recommendations for JavaScript, Python, Java, TypeScript, Ruby, Golang and many other programming languages and accept, reject or manually edit them. Copilot will adapt to the edits developers make, matching particular coding styles to autofill boilerplate or repetitive code patterns and recommend unit tests that match implementation code.
Copilot extensions will be open for JetBrains and Noevim in addition to VS Code or in the cloud on GitHub Codespaces. One new feature coinciding with the general release of Copilot is Copilot Explain, which translates code into natural language descriptions. The goal at the end of this is to help novice developers or those working with an unfamiliar codebase.
Due to the complicated nature of AI models, Copilot remains an imperfect system. GitHub warns that it can produce insecure coding patterns, bugs and references to outdated APIs, or idioms that reflect the less-than-perfect code in its training data. The code that Copilot suggests may not always compile, run, or even make sense because it doesn’t really test the suggestions. Additionally, on rare occasions, Copilot suggestions may include personal information, such as names and emails, verbatim from the training set — and worse, “biased, discriminatory, abusive, or abusive” text.
Do You Know About This: GitHub Copilot Generated Insecure Code In 40% Of Circumstances During Experiment
GitHub said that it’s implemented filters to block emails when shown in standard formats, and offensive words, and that it’s in the process of building a filter to help detect and suppress code that’s repeated from public repositories. “While we are working hard to make Copilot better, code suggested by Copilot should be carefully tested, reviewed, and vetted, like any other code,” the disclaimer on the Copilot website reads.
While Copilot has supposedly improved since its launch in technical preview last year, it’s unclear how much. The capabilities of the underlying Codex model – a descendant of OpenAI’s GPT-3 – have since been matched (or even surpassed) by systems such as DeepMind’s AlphaCode and the open source PolyCoder.
“We are seeing progress in Copilot generating better code, We’re using our experience with [other] tools to improve the quality of Copilot suggestions — e.g., by giving extra weight to training data scanned by CodeQL, or analyzing suggestions at runtime,” Salva asserted — “CodeQL” referring to GitHub’s code analysis engine for automating security checks. “We’re committed to helping developers be more productive while also improving code quality and security. In the long term, we believe Copilot will write code that’s more secure than the average programmer.”
The lack of transparency doesn’t seem to have dampened enthusiasm for Copilot, which according to Microsoft today suggests that about 35% of the code in languages like Java and Python was generated by the developers in the technical preview. Tens of thousands have used the tool regularly during the preview, the company claims.
Do You Know About This: GitHub Copilot Is ‘Unacceptable And Unjust’ Says Free Software Foundation
Apart from this announcements, Github Copilot anchored a series of product rollouts unveiled at Microsoft Build. Other rollouts announced at Build include:
- Microsoft Dev Box, a cloud service providing developers with secure, ready-to-code workstations for hybrid teams of any size. Teams preconfigure Dev Boxes for specific projects and tasks. Unified management, security, and compliance are maintained by IT via Windows 365. Developers can sign up for a private preview of Dev Box.
- A preview of Microsoft Power Pages, a tool for building secure, low-code webpages. The tool is part of the Microsoft Power Platform.
- Microsoft’s Azure Container Apps, for running microservices and containerized applications on a serverless platform, has reached general availability.
- Microsoft Intelligent Data Platform, a cloud data platform that integrates databases, analytics, and governance.
- Project Volterra, a development kit with AI capabilities and a comprehensive Arm-native developer toolchain.
Let’s see when this tool will be available for all.