{"id":"arxiv/2604.16424","url":"https://arxiv.org/abs/2604.16424","tags":["cs.CR","cs.AI","cs.CL","cs.LG","math.OC"],"score":0,"title":"Safety, Security, and Cognitive Risks in State-Space Models: A Systematic Threat Analysis with Spectral, Stateful, and Capacity Attacks","domain":"cs.CR","source":"arxiv","authors":["Manoj Parmar"],"reviews":0,"subject":"cs.CR","summary":"State-Space Models (SSMs) -- structured SSMs (S4, S4D, DSS, S5), selective SSMs (Mamba, Mamba-2), and hybrid architectures (Jamba) -- are deployed in safety-critical long-context applications: genomic analysis, clinical time-series forecasting, and cybersecurity log processing. Their linear-time scaling is compelling, yet the security properties of their compressed-state recurrent architectures remain unstudied. We present the first systematic treatment of SSM safety, security, and cognitive risks. Seven contributions: (1) Formal threat framework -- SSM Attack Surface (five layers), State…","category":"research","comments":0,"disputed":false,"division":"cs","mentions":0,"discussed":0,"published":"2026-04-04","reactions":{}}