Hyperscience, a leader in enterprise AI infrastructure software specializing in Intelligent Document Processing (IDP), has launched Hypercell for SNAP, a solution designed to assist U.
S. State and Local Governments in meeting increased compliance mandates for the Supplemental Nutrition Assistance Program (SNAP) under H.
R. 1.
Following the enactment of H.
R. 1, states administering SNAP benefits face significant operational challenges and increased administrative burden. An estimated 40% of SNAP applications are rejected due to issues like incompleteness, illegibility, or incorrect filling. This high rejection rate highlights deficiencies in an application process that can be confusing and inaccessible, particularly for vulnerable populations facing economic hardship, language barriers, or limited digital literacy.
The new legislation mandates that the 42 million SNAP beneficiaries nationwide must reapply or recertify eligibility every six months, effectively doubling the paperwork for state agencies already managing overwhelmed systems. Additionally, new work requirements will further increase document volume and complexity. These expanded eligibility compliance requirements take effect in October 2026, with federal penalties for states exceeding a 6 percent SNAP Payment Error Rate (PER) beginning in October 2027. Currently, 44 states surpass this 6 percent error rate, exposing them to significant fines and potential risks to access the $100 billion in federal funding for SNAP programs.
Andrew Joiner, CEO of Hyperscience, stated that while state governments have invested heavily in legacy technologies that have not resolved the core SNAP challenge, the fundamental issue lies with document processing. He emphasized that documents remain the primary bottleneck, causing application delays and exposing states to penalties.
Hypercell for SNAP is presented as a scalable, modular solution developed to manage the operational complexities of SNAP. This includes processing over 30 types of eligibility documents per applicant, which often feature varying formats and handwriting issues across millions of applicants. Documents include driver’s licenses, state IDs, passports, lease agreements, utility bills, mortgage statements, pay stubs, employer letters, tax returns, child care receipts, bank statements, and vehicle registrations.
The solution utilizes Hyperscience’s proprietary Optical Reasoning and Cognition Agent (ORCA), a Vision Language Model (VLM) framework. ORCA is designed for high-variability document processing in SNAP applications, offering rapid document classification and extraction with minimal setup, automated validation of required documents, and a streamlined human review interface. Its advanced image processing capabilities aim to deliver enhanced clarity and extraction accuracy, addressing issues that legacy Optical Character Recognition (OCR) solutions often struggle with. By deploying this AI-driven solution, states can potentially reduce payment errors by 50 percent, cut administrative costs, and significantly accelerate SNAP benefit delivery from an average of 26 days to approximately seven days.
Hyperscience emphasizes the security and compliance of Hypercell for SNAP. The solution is FedRAMP® High authorized, achieved in partnership with Palantir Technologies through its FedStart Program. It also holds TX-RAMP Level 2 certification for secure data processing in Texas. Hyperscience undergoes annual independent SOC 2 Type II audits and maintains Cyber Essentials Plus certification. The platform offers deployment flexibility across public cloud, on-premises, or air-gapped environments.
While optimized for SNAP, over 85% of the Hypercell platform’s capabilities are applicable to other entitlement programs, including Medicaid, TANF, and LIHEAP, offering a pathway for broader government modernization.
