Abstract—Architectural decay can manifest as evolution of architectural smells, degrading integrity, and increasing maintenance costs. Existing techniques capture smells post hoc, or predict on component level, acting too late or on a too coarse granularity. We investigate if the risk of introducing architectural smells can already be predicted when issues are opened. Thus, we propose an issue-level prediction approach that utilizes the semantic representations of Large Language Models (LLMs). To enable training and evaluation, we construct a multi-project dataset by linking issues to smells via smell-inducing changes. On this dataset, we train classifiers to identify high-risk issues and conduct an empirical study comparing seven different representations and nine classifiers. Our best performing classifier (SVM with OpenAI embeddings) achieves F1-scores of up to 0.506, with recall of about 0.74. This means that our approach can identify approximately 74% of smell-inducing issues before implementation begins. When design alternatives are still being considered. our approach is able to give early warnings about potential architectural risks. This work shifts from reactive remediation to preventive quality assurance, reducing the accumulation of architectural technical debt.