The challenge of systematically modifying and optimizing regulatory elementsfor precise gene expression control is central to modern genomics and syntheticbiology. Advancements in generative AI have paved the way for designing syntheticsequences with the aim of safely and accurately modulating gene expression.We leverage diffusion models to design context-specific DNA regulatorysequences, which hold significant potential toward enabling novel therapeutic applicationsrequiring precise modulation of gene expression. Our framework usesa cell type-specific diffusion model to generate synthetic 200 bp DNA regulatoryelements based on chromatin accessibility across different cell types. We evaluatethe generated sequences based on key metrics to ensure they retain properties ofendogenous sequences: transcription factor binding site composition, potential forcell type-specific chromatin accessibility, and capacity for sequences generated byDNA diffusion to activate gene expression in different cell contexts using state-ofthe-art prediction models. Our results demonstrate the ability to robustly generateDNA sequences with cell type-specific regulatory potential. DNA-Diffusion pavesthe way for revolutionizing a regulatory modulation approach to mammalian syntheticbiology and precision gene therapy.