May 01, 2025 11:04:00

Around 30% of Microsoft's product code is written by AI.

Microsoft CEO Satya Nadella has revealed that 20 to 30 percent of the code in the company's software products is written by artificial intelligence (AI).

Microsoft's CEO reveals that AI writes up to 30% of its code — some projects may have all of its code written by AI | Tom's Hardware

https://www.tomshardware.com/tech-industry/artificial-intelligence/microsofts-ceo-reveals-that-ai-writes-up-to-30-percent-of-its-code-some-projects-may-have-all-of-its-code-written-by-ai

At LlamaCon 2025 on April 29, 2025, CEO Nadella and Meta CEO Mark Zuckerberg discussed the development of AI and contributions to the open source ecosystem. In the discussion, CEO Nadella revealed that even large companies like Microsoft are heavily dependent on AI.

During the discussion, Zuckerberg and Nadella detailed the scale of their companies' AI adoption, with Nadella revealing that one-third of the code in Microsoft's repositories and projects is now generated by AI.

Meanwhile, Zuckerberg did not provide specific figures on Meta's AI usage, but outlined plans for future AI models to build and create future iterations of the Llama AI model.

Nadella claims that AI can generate better code in Python than in

C++ . The reason why AI writes better code in Python is because Python has advantages such as simple syntax, dynamic typing style, error handling, and memory management. On the other hand, C++ and C languages do not have native garbage collectors . In addition, technology media Tom's Hardware points out that C and C++ are mainly related to low-level code, making automation difficult.

'The fact that a big company like Microsoft is so reliant on AI is a clear indication of how AI is revolutionizing software development. But it also raises the level of anxiety and uncertainty for new software developers,' Tom's Hardware reports.

During the third quarter 2024 earnings call, Google CEO Sundar Pichai revealed that '25% of Google's new code is written by AI.' At the time of writing, AI is primarily used to enhance repetitive, data-intensive, and predictable tasks. This has reduced entry-level work and significantly increased the company's efficiency. While it is true that AI-generated code has improved significantly in recent years, 'it still requires oversight by senior developers to avoid disrupting the production environment,' Tom's Hardware points out.

In addition, Tom's Hardware said, 'It is natural that this situation will cause concern about job losses for new programmers, especially in today's competitive job market. However, it is unwise for software developers to ignore or rely too heavily on AI. A balanced approach with a solid understanding of the fundamentals, an understanding of how to use AI as a tool, and good critical thinking skills seems to be the best path. It is difficult to predict the future, but this ratio will probably change. '