TLDR: Google has issued new internal guidance for its software engineers on leveraging AI for coding, emphasizing productivity gains while maintaining human oversight in critical areas like code review, security, and maintenance. This move follows a period of internal experimentation and comes as over 30% of Google’s code is now AI-generated, a figure that has increased by over 5% in six months.
Google has officially released new internal guidance for its software engineers regarding the use of artificial intelligence in coding. This comprehensive document, developed by a group of Google’s own software engineers, aims to boost productivity, enable faster iteration, and enhance product development across the company. The guidance was disseminated following an email to all software engineers on Monday, outlining recommendations and best practices for integrating AI into their daily workflows.
The tech giant is actively encouraging its developers to adopt AI tools, with CEO Sundar Pichai stating in April 2025 that over 30% of Google’s code is now AI-generated. This represents a significant increase of over 5% in just six months and involves AI producing code that is then reviewed and accepted by human engineers. Pichai also noted earlier this month that AI usage has contributed to an estimated 10% increase in engineering velocity, signifying a greater amount of software work accomplished within a given period.
Despite the push for AI adoption, the guidance strongly emphasizes the continued importance of human oversight. Key areas such as code review, security protocols, and ongoing maintenance are highlighted as requiring rigorous human involvement. Google’s engineering teams are already extensively using AI tools for coding, fixing bugs, and various other tasks, and these new recommendations are designed to help all software engineers maximize the benefits of AI in their work.
Also Read:
- Google Introduces AI-Powered “Web Guide” to Enhance Search Relevance
- Goldman Sachs Employees Generate Over a Million AI Prompts Monthly, Signaling Widespread Adoption
The guidance also covers the capabilities and limitations of current AI technologies and aims to help engineers understand how to build AI-based systems. Specifically, tech leads and managers are expected to gain proficiency in incorporating AI development into their teams. This internal initiative aligns with Google’s external releases over the past year, including Gemini Code Assist, Gemini CLI, and Gemini in Android Studio, showcasing a company-wide commitment to integrating AI into its development ecosystem.
