SafeSeal is an AI-driven watermarking technology designed to protect large language model (LLM) outputs from unauthorized use while preserving text quality and readability. The system embeds detectable watermarks through context-aware synonym substitutions while using named entity recognition to preserve critical information and maintain semantic integrity. Performance testing demonstrated a BERTScore of 0.981, an entity similarity score of 0.962, and a watermark detection rate of 95.1%, indicating strong content fidelity and reliable verification. Potential applications include intellectual property protection, AI content authentication, compliance, and secure deployment of commercial LLMs. The patent-pending technology is at TRL 3 and available for licensing.

GrantsMate is an AI-powered research support platform developed at the University at Albany that streamlines key research activities by integrating funding discovery, collaborator identification, and institutional policy guidance into a single system. Using machine learning, natural language processing, retrieval-augmented generation, and vector databases, the platform delivers personalized, context-aware recommendations through a conversational interface. Its modular architecture supports both cloud and on-premises deployment, allowing institutions to tailor the system to their needs. By reducing inefficiencies created by fragmented research support tools, GrantsMate helps researchers identify funding opportunities, build collaborations, and navigate administrative requirements more effectively. The patent-pending technology is at TRL 3 and available for licensing.

AgentHeart is an AI-powered platform designed to support continuous cardiovascular monitoring and clinical decision-making. Unlike traditional “black-box” AI tools, AgentHeart emphasizes transparency and collaboration, functioning as a trusted assistant that continuously analyzes patient cardiovascular data and provides real-time, actionable insights to clinicians and care teams. By monitoring trends and identifying potential risks as they emerge, the platform supports earlier interventions, reduces diagnostic delays, and enhances care coordination. Potential applications include cardiology clinics, hospital cardiac units, telemedicine, remote patient monitoring, and clinical research. The patent-pending technology is at TRL 3 and available for licensing.

This AI-Driven System for Integrated Patent Valuation, Marketability Assessment, and Prior-Art Intelligence uses artificial intelligence, semantic analysis, and market intelligence to help inventors and intellectual property professionals evaluate patent originality, commercial potential, and competitive positioning. The platform transforms patent records into searchable vector embeddings, enabling advanced prior-art discovery and similarity analysis, while a Product-Market Fit (PMF) scoring engine assesses potential market value and relevance. By combining technical patent evaluation with market insights in a single workflow, the system simplifies complex decision-making and supports patent filing, licensing, portfolio management, and commercialization strategies. The patent-pending technology is at TRL 3 and available for licensing.

MyQL: The Retrieval Augmented Tutor is an AI-powered educational platform designed to enhance learning by integrating course materials directly into a personalized tutoring experience. Using retrieval-augmented generation (RAG), the system provides context-aware answers with citations, interactive practice opportunities, and tailored feedback grounded in instructor-provided content. MyQL promotes critical thinking by encouraging students to evaluate and correct information rather than passively consume answers, while also supporting experiential and collaborative learning. Designed to operate on low-end hardware, it prioritizes accessibility, adaptability, and privacy. Potential applications include higher education, online learning, workforce training, and AI literacy programs. The patent-pending technology is at TRL 4 and available for licensing.

With the widespread application of radar in space target detection and identification, acquiring radar images of non-cooperative targets at different attitudes has become the data foundation for supporting deep learning-based automatic target recognition algorithms. However, the uncontrollable motion trajectories of non-cooperative targets and limited illumination time render the acquisition of measured images extremely difficult; although anechoic chamber measurements and electromagnetic computations can obtain multi-attitude images, they are respectively constrained by high costs and computational time, making it difficult to meet the demand for large-scale data. Existing image generation methods based on Generative Adversarial Networks (GANs) can augment data scales, but most can only control the azimuth angle without precisely constraining the elevation angle, resulting in generated images whose projection morphology does not match reality. Therefore, how to achieve radar image generation with precise dual-dimensional control over both azimuth and elevation angles has become a key technical challenge in data augmentation for radar recognition of non-cooperative targets.