Vibe coding, a term coined by Andrej Karpathy in February 2025, describes a development approach where users fully embrace AI-generated code from large language models like Cursor Composer w Sonnet. Proponents argue this methodology democratizes software creation, enabling non-experts to build applications. For example, a veterinary technician used vibe coding to create an app for tracking her senior cat's insulin shots.

However, the practice faces significant criticism regarding software security and maintainability. Researchers from Georgia Tech's School of Cybersecurity and Privacy analyzed 43,000 security advisories and traced 74 vulnerabilities directly to AI-generated code, 14 of which were deemed critical. The team estimates the actual number of vulnerabilities is five to ten times higher due to underreporting.

Despite these risks, adoption among professionals is surging. A survey of 1,100 programmers found 72 percent use AI coding tools daily, with approximately 42 percent of their codebase now AI-generated or assisted. Experts predict AI will account for more than half of all code by next year.

This shift raises concerns about the future of junior developer roles, as traditional entry-level coding positions diminish. The distinction between fully AI-generated code and AI-assisted development remains crucial for understanding the evolving landscape of software engineering.