- Google survey finds 90% of developers use AI at work, 80% see higher productivity
- Two-thirds rely heavily on AI, many use it 2 hours/day
- Many still don’t fully trust AI’s output, suggesting more needs to be done
It’s no secret that developers are using AI to help with their repetitive coding activities, but new Google[1] research[2] has revealed the true extent of AI use – two in three (65%) software devs saying they now heavily rely on AI tools[3].
In terms of general use, nearly all development professionals (90%) reportedly they now use AI, a sharp rise over the 14% observed in 2024.
Today, the median time spend using AI in workflows among these type of workers now stands at two hours.
Google says developers are really using AI
The benefits for developers using the technology are clear – four in five agree they see higher productivity from AI use, and three in five (59%) claim to see improvements in their code quality.
When using generative AI, developers can free themselves up more time for problem-solving, design and oversight, with AI handling the lower-level coding demands.
However, there remains some resentment over handing over work to computer intelligence, with fewer than a quarter (24%) trusting AI outputs ‘a lot’ or ‘a great deal’. As such, developers tend to see AI as a supportive tool, and not a replacement for human judgement.
Despite our worst fears, the report also reveals that AI adoption has not significantly changed how developers experience their work lives yet, with programmers feeling that their expertise is still valued despite the rise of time-saving AI.
For companies, AI has both positive and negative implications, with Google Senior Director for Product Management Ryan J Salva calling the tech a “mirror and a multiplier.”
“In cohesive organizations, AI boosts efficiency. In fragmented ones, it highlights weaknesses,” Salva noted[4].
The decade-old DevOps Research and Assessment (DORA) program split workers into seven team archetypes to dive a little deeper into how AI affects developers at work, concluding that organizations must “evolve their culture, processes and systems to support a new era of software development.”