- Google issues advice developed on the engineer on the use of AI for coding
- Approximately a third of the company code is now generated by AI
- Other companies also use more AI to code
Google has published new advice on use and best practices for its internal developers using a generative AI to help them code, and it may not be surprising that the technology giant pushes the agenda.
After a period of experimentation with AI and to determine its use cases, the document was created by engineers (via 9TO5GOOGLE) Before being approved and broadcast.
The advice largely encourages developers to adopt AI to stimulate productivity, allow faster iteration and improve product development, but the importance of humans in the maintenance of security is underlined.
Google shares AI coding directives with workers
The directives would be covering the use of AI for coding assistance, its limits and the construction of AI -based systems. However, Google also noted that he wanted to see workers adopting AI in other regions of their work to help further stimulate productivity.
Nevertheless, humans clearly remain an integral part of the development process, with the code examination, security and maintenance of all areas that require additional work.
“Our engineering teams already regularly use our AI tools for coding, correction of bugs, and much more,” confirmed the company.
“These recommendations and best practices, developed by a group of Google software engineers, are designed to help all SWE through the company to get the most out of AI in their daily work.”
The push comes in the middle of the growing adoption of Genai on the scale of the software engineering industry. Earlier in 2025, the CEO of Google Sundar Pichai revealed that around a third of the company code was now generated by AI.
Other companies, including Microsoft and Meta, also increase the amount of code generated by the AI they use, and a recent study revealed that 30.1% of us the Python code on Github were generated by AI.
Google’s most formal advice on AI in coding to date could also inform policies of other companies. It also highlights the growing dependence of the company with regard to generative AI tools.
