AI now writes over 25% of Google’s new code—yep, one in four lines of fresh, shiny code comes from machines, not humans. Sundar Pichai recently broke it down in Google’s earnings call, showing how AI is becoming a powerhouse in software development. It’s all supervised, of course, by engineers guiding this digital workforce, turbocharging productivity and efficiency while also letting teams get more done faster.
A quarter of Google’s code isn’t small potatoes, and Google isn’t alone. According to Stack Overflow, more than three-quarters of developers worldwide are already using or planning to use AI to help with software development. GitHub’s survey reveals even higher numbers: 92% of US developers are already harnessing AI coding tools.
But here’s the rub: AI isn’t just taking on tasks; it’s quietly changing the landscape. As AI cranks out more code, the human skill base could shrink. What happens when the errors start popping up, glitches buried deep in code written by AI that itself was born from AI? It’s a recipe for an endless loop of confusion—think an AI-driven bug hunt where humans struggle to keep up.
And we’re just warming up. AI coding tools began their rocket-fueled journey in 2022 with GitHub’s Copilot. Since then, Meta, Google, OpenAI, and Anthropic have all jumped on board, releasing their own AI-powered coding platforms. GitHub even stepped it up recently, opening Copilot to work with models from Anthropic, Google, and OpenAI. In other words, AI is coding the future, and it’s moving faster than we think.
Image: DIW-Aigen
Read next: Study Finds Age Assurance Methods Fail to Protect Children from Online Harm and Privacy Risks
A quarter of Google’s code isn’t small potatoes, and Google isn’t alone. According to Stack Overflow, more than three-quarters of developers worldwide are already using or planning to use AI to help with software development. GitHub’s survey reveals even higher numbers: 92% of US developers are already harnessing AI coding tools.
But here’s the rub: AI isn’t just taking on tasks; it’s quietly changing the landscape. As AI cranks out more code, the human skill base could shrink. What happens when the errors start popping up, glitches buried deep in code written by AI that itself was born from AI? It’s a recipe for an endless loop of confusion—think an AI-driven bug hunt where humans struggle to keep up.
And we’re just warming up. AI coding tools began their rocket-fueled journey in 2022 with GitHub’s Copilot. Since then, Meta, Google, OpenAI, and Anthropic have all jumped on board, releasing their own AI-powered coding platforms. GitHub even stepped it up recently, opening Copilot to work with models from Anthropic, Google, and OpenAI. In other words, AI is coding the future, and it’s moving faster than we think.
Read next: Study Finds Age Assurance Methods Fail to Protect Children from Online Harm and Privacy Risks
