In recent months, I’ve noticed a troubling trend with AI data-linked-post=”2671645555″ href=”https://spectrum.ieee.org/vibe-coding” target=”_blank”>coding assistants. After two years of steady improvements, over the course of 2025, most of the core models reached a quality plateau, and more recently, seem to be in decline. A task that might have taken five hours assisted by AI, and perhaps ten hours without it, is now more commonly taking seven or eight hours, or even longer. It’s reached the point where I am sometimes going back and using older versions of data-linked-post=”2672500550″ href=”https://spectrum.ieee.org/large-language-model-performance” target=”_blank”>large language models (LLMs).
I…
