GOOGLE: AI NOW WRITES 75% OF NEW CODE

Google disclosed that AI now accounts for 75 percent of newly written code across the company, a substantial increase in machine-assisted development. The remaining 25 percent comes from human developers, who handle code review and quality assurance for all AI-generated output. The figure reflects the broader industry trend of integrating AI coding tools into development workflows. Google has invested heavily in AI-powered code generation through products like Codex and its own internal systems. The company does not specify which AI models power this code generation or whether the 75 percent figure represents all engineering teams uniformly. The metric also does not indicate whether AI-written code requires more revision or produces different quality levels compared to human-written alternatives. This development comes as major tech firms increasingly deploy AI for software development tasks. Similar tools from competitors like GitHub Copilot and Microsoft have gained adoption among developers seeking to accelerate coding processes. Google's disclosure raises questions about code review depth, security protocols, and the long-term trajectory of human developer roles. The company emphasizes that human review remains mandatory, suggesting it views AI-generated code as requiring the same scrutiny as human-written code. The implications extend beyond productivity metrics. Relying on AI for three-quarters of code creation could affect how developers learn, maintain codebases, and catch subtle bugs. It may also shift demand in the tech labor market toward code review and architecture roles rather than direct implementation work. Google has not released specifics on performance metrics, bug rates, or security outcomes comparing AI-generated versus human-written code at scale. The company also has not addressed whether clients and partners are informed when their products contain substantial AI-generated components.