ChatGPT on macOS Now Supports Direct Code Editing
Mac users can now take advantage of a powerful new feature in the ChatGPT app—direct code editing in supported development tools such as Xcode, VS Code, and JetBrains. This update also introduces an optional "auto-apply" mode, allowing ChatGPT to make edits automatically without requiring additional clicks.
Availability and Future Updates
This feature is currently available to subscribers of ChatGPT Plus, Pro, and Team, who can enable it by updating their macOS app. OpenAI has announced that Enterprise, Edu, and free-tier users will gain access to the functionality next week.
Alexander Embiricos, a member of OpenAI’s product team for desktop applications, shared on X that a similar feature will be coming to the ChatGPT Windows app soon.
Enhanced AI Coding Assistance
The direct code editing capability builds upon OpenAI’s “work with apps” feature, which was introduced in beta in November 2024. This functionality allows ChatGPT to read code directly within select developer environments, reducing the need to manually copy and paste code into the chatbot.
Competing with AI Coding Tools
With this enhancement, ChatGPT positions itself as a stronger competitor to popular AI-powered coding assistants such as Cursor and GitHub Copilot. OpenAI is also rumored to be working on a dedicated product specifically designed for software engineers, expected to launch in the coming months.
AI Coding Tools Are on the Rise
AI coding assistants are growing rapidly in popularity. A recent GitHub survey revealed that the vast majority of developers are now incorporating AI tools into their workflows. Jared Friedman, a partner at Y Combinator, recently claimed that a quarter of YC’s latest startup batch (W25) has 95% of their codebases generated by AI.
Challenges and Risks
However, AI-assisted coding comes with challenges, including security vulnerabilities, copyright concerns, and reliability issues. A study by software vendor Harness found that developers spend more time debugging AI-generated code and addressing security flaws compared to human-written contributions. Meanwhile, a Google report indicated that while AI can accelerate code reviews and improve documentation, it may also compromise software delivery stability.