For many developers, AI coding tools like GitHub Copilot, Claude Code, and Cursor have become an indispensable part of their daily workflow. While these technologies speed up processes like boilerplate code generation, test writing, and debugging, they also bring a significant risk of addiction. Quentin Rousseau, CTO and co-founder of Rootly, describes this new situation created by AI-powered coding processes as a kind of work addiction. Rousseau states that he can’t help but watch the AI ​​revise the code even late at night, and this disrupts his sleep patterns.

AI-Related Work Addiction and Burnout

The rapid feedback loop offered by AI tools can become a source of both dopamine and adrenaline for developers. However, this situation makes it difficult for software developers to manage their working hours, significantly disrupting their work-life balance.

According to the research, 43% of developers continue to code with AI even after working hours. 32% prefer to stay in this cycle by sacrificing sleep hours.

AI tools make it difficult for developers to detach from their work, leading to 39% more mental fatigue. This is considered a harbinger of a new type of burnout syndrome in the software world.

Although the use of artificial intelligence increases career opportunities such as promotion or salary increases for some developers, 51% state that this intense pace causes burnout. Constantly being stuck in a feedback loop prevents developers from completing their work and leaving their computers.

The Obligation to Verify and Declining Trust

The 2025 Stack Overflow Developer Survey raises significant questions about the accuracy of AI-generated responses. 45% of respondents expressed discomfort with the process of filtering out the almost accurate but flawed answers provided by AI.

Although the adoption rate of AI tools has reached 80%, trust in these tools has decreased compared to previous years. The trust rate, which was previously 40%, has fallen to 29% this year.

Developers’ positive view of AI has also decreased from 72% to 60% year-on-year. Obtaining quick output creates a new obligation on developers called the obligation to verify.

It remains the responsibility of developers to verify the accuracy, security, and system architecture compliance of code generated by AI. Employers viewing AI as a capacity-enhancing tool creates pressure on employees to develop more features and close more tickets.

This eliminates the time savings provided by AI, forcing developers to conduct more reviews and verifications. As a result, speeding up each step of the software production pipeline can lead to a more intense and tiring work process.

Do you think AI coding tools truly increase your work efficiency, or do they negatively impact your work system?

Do AI coding tools create addiction among programmers? Research shows that 10% of developers are addicted to these tools.