<p>Diabetes-associated cognitive decline (DACD) represents a complex disorder arising from the intersection of metabolic dysfunction, neuroinflammation, and neurodegenerative signaling. Despite increasing clinical recognition, the molecular network governing DACD remains poorly defined, and a systems-level network pharmacology approach has not previously been applied to this condition. Curcumin–indole-3-propionic acid conjugate (CUR-IPA), developed from two bioactive neuroprotective scaffolds, offers a rational multi-target framework to modulate DACD-associated pathways. DACD-related genes were compiled and analyzed using protein–protein interaction networks, hub gene identification, and pathway enrichment analysis. Drug-likeness and ADMET properties of CUR, IPA, and CUR-IPA were predicted in silico. Molecular docking was performed against prioritized targets, followed by 200-ns molecular dynamics simulations. Protein–ligand stability and dynamics were assessed using RMSD, RMSF, DCCM, elastic network model, and Markov state model analyses. Binding energetics were quantified using MM-PBSA calculations with residue-level decomposition. Network pharmacology identified PSEN1 and TGFβ1 as central hub proteins linking amyloid regulation, neuronal maintenance, immune signaling, and metabolic pathways in DACD. CUR-IPA satisfied multiple drug-likeness filters and exhibited favorable pharmacokinetic and toxicity profiles. Docking and MD simulations demonstrated stable binding of CUR-IPA within the active sites of PSEN1 and TGFβ1. ENM revealed restricted collective motions around ligand-binding regions, while MSM analysis identified energetically favorable and kinetically stable conformational states. MM-PBSA calculations confirmed strong binding affinities driven by key catalytic residues. This study establishes the first network pharmacology framework for DACD and demonstrates that CUR-IPA acts as a stable, multi-target modulator of key neurodegenerative and metabolic pathways.</p>