Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Olga A. Shvetsova has a total of 46 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Decoupling Intelligence from Governance: A Dynamic Bilateral Architecture for Real-Time Enterprise AI Compliance

    by Danila Katalshov, Olga Shvetsova, Sang-Kon Lee, Sviatlana Koltun

    The widespread adoption of Generative Artificial Intelligence (GenAI) in regulated enterprises is currently hindered by the “Static Alignment Trap”: the inability of traditional safety methods, such as Reinforcement Learning from Human Feedback (RLHF), to adapt to rapidly shifting compliance landscapes without costly retraining. This paper introduces and evaluates the Agreement Validation Interface (AVI), a modular governance architecture that functions as a deterministic middleware layer. By decoupling governance from the core inference engine, AVI implements Dynamic Bilateral Alignment (DBA), enforcing policy constraints at both the input and output stages through vector-based semantic retrieval. Adopting a Design Science Research (DSR) methodology, we validated the system against the FinanceBench financial benchmark (N=150 queries, three repeated runs, 450 total observations) and a proprietary Russian-language provocative content dataset developed internally at MWS AI (N=201 queries; not publicly available). The empirical results demonstrate that the architecture achieves an 83.2% Large Language Model (LLM)-judge compliance rate (95% confidence interval, CI: 79.4–87.1%), statistically significantly exceeding the unfiltered baseline of 63.7% (Δ=+19.5 percentage points (pp), t=4.02, p=0.002). The vector-based input filter achieves perfect detection performance (Precision =1.000, Recall =1.000, F1 =1.000). Cross-domain validation on 201 Russian-language provocative queries confirms generalizability (Recall =0.985, LLM compliance among triggered queries =0.977). The operational Time-to-Compliance for enforcing new rules was reduced from hours (model fine-tuning) to under five seconds (vector indexing). These findings suggest that enterprise AI safety requires an architectural shift from model-centric training to system-centric control, complemented by system-prompt-level anti-inference governance. We conclude that AVI offers a scalable, cost-effective, and statistically validated framework for auditable AI compliance, independent of the underlying model provider.

  • Top Cited Papers
    Open Access

    Innovative Guardrails for Generative AI: Designing an Intelligent Filter for Safe and Responsible LLM Deployment

    by Olga Shvetsova, Danila Katalshov, Sang-Kon Lee

    This paper proposes a technological framework designed to mitigate the inherent risks associated with the deployment of artificial intelligence (AI) in decision-making and task execution within the management processes. The Agreement Validation Interface (AVI) functions as a modular Application Programming Interface (API) Gateway positioned between user applications and LLMs. This gateway architecture is designed to be LLM-agnostic, meaning it can operate with various underlying LLMs without requiring specific modifications for each model. This universality is achieved by standardizing the interface for requests and responses and applying a consistent set of validation and enhancement processes irrespective of the chosen LLM provider, thus offering a consistent governance layer across a diverse LLM ecosystem. AVI facilitates the orchestration of multiple AI subcomponents for input–output validation, response evaluation, and contextual reasoning, thereby enabling real-time, bidirectional filtering of user interactions. A proof-of-concept (PoC) implementation of AVI was developed and rigorously evaluated using industry-standard benchmarks. The system was tested for its effectiveness in mitigating adversarial prompts, reducing toxic outputs, detecting personally identifiable information (PII), and enhancing factual consistency. The results demonstrated that AVI reduced successful fast injection attacks by 82%, decreased toxic content generation by 75%, and achieved high PII detection performance (F1-score ≈ 0.95). Furthermore, the contextual reasoning module significantly improved the neutrality and factual validity of model outputs. Although the integration of AVI introduced a moderate increase in latency, the overall framework effectively enhanced the reliability, safety, and interpretability of LLM-driven applications. AVI provides a scalable and adaptable architectural template for the responsible deployment of generative AI in high-stakes domains such as finance, healthcare, and education, promoting safer and more ethical use of AI technologies.


  • Open Access

    Managing Global Talent: Innovative Solutions and a Sustainable Strategy Using a Human-Centric Approach

    by Olga Shvetsova

    This study examines the challenges and opportunities in global talent management (GTM) systems within developed economies, with a specific focus on Republic of Korea. The research investigates both internal and external factors—positive and negative—that shape GTM strategies across key national high-tech industries. Framed within a human-centric model, the study identifies three dominant global megatrends—technological, socio-economic, and cultural—that significantly influence talent ecosystem dynamics. The study is framed as a qualitative interview-based analysis, supported by quantitative descriptive data for contextual analysis. A qualitative approach was employed, addressing the quantitative analysis of statistical labor market data to qualitative interviews (n = 42) conducted with high-level STEM professionals pursuing global talent status. Quantitative statistics indicate a measurable decline in global talent inflow to Republic of Korea in the past year. Qualitative findings reveal critical deterrents—including regulatory, organizational, and cultural barriers—as well as motivational drivers influencing global mobility decisions. These factors were synthesized within the three identified megatrends to provide a comprehensive view of GTM system dynamics. Based on the findings, the study offers strategic, human-centered recommendations for enterprises and academic institutions to enhance their GTM practices in an era of accelerating technological change. The results offer practical guidance for policymakers, business leaders, educators, and researchers seeking to align talent strategies with evolving global trends.


  • Open Access

    Innovative Guardrails for Generative AI: Designing an Intelligent Filter for Safe and Responsible LLM Deployment

    by Olga Shvetsova, Danila Katalshov, Sang-Kon Lee

    This paper proposes a technological framework designed to mitigate the inherent risks associated with the deployment of artificial intelligence (AI) in decision-making and task execution within the management processes. The Agreement Validation Interface (AVI)—a modular, LLM-agnostic API gateway - has been designed to enhance the trustworthiness and governance of generative AI systems. AVI facilitates orchestration of multiple AI subcomponents for input-output validation, response evaluation, and contextual reasoning, thereby enabling real-time, bidirectional filtering of user interactions. A proof-of-concept (PoC) implementation of AVI was developed and rigorously evaluated using industry-standard benchmarks. The system was tested for its effectiveness in mitigating adversarial prompts, reducing toxic outputs, detecting personally identifiable information (PII), and enhancing factual consistency. Results demonstrated that AVI reduced successful fast injection attacks by 82%, decreased toxic content generation by 75%, and achieved high PII detection performance (F1 score ≈ 0.95). Furthermore, the contextual reasoning module significantly improved the neutrality and factual validity of model outputs. Although the integration of AVI introduced a moderate increase in latency, the overall framework effectively enhanced the reliability, safety, and interpretability of LLM-driven applications. AVI provides a scalable and adaptable architectural template for the responsible deployment of generative AI in high-stakes domains such as finance, healthcare, and education, promoting safer and more ethical use of AI technologies.


  • Promoting Sustainable Development Goals Through Green Campuses

    by Hein Thiha, Olga Shvetsova

    According to a framework provided by United Nations' Sustainable Development Goals for a more sustainable future, there are 17 SDGs in total. Among them, SDG 15 and SDG 4 are addressing the various aspects of sustainable development for promoting quality education and lifelong learning by considering green campus concept for sustainability in near future. Green campus serves as living laboratory for sustainable practices and contribute to multiple SDGs by promoting environmental stewardship, fostering educational awareness, and innovation. The findings of this study could aid how university are initiating the green campus through modern technology and ecological potential of green zone around the campus with specialized focusing on Universities in South Korea. Through this initiative on fostering partnership with government agencies and local organization, university administration will support the students to engage in promoting sustainability on campus giving a sense of community collective responsibility for environmental awareness.

  • Open Access

    Behavioral Model Deployment for the Transportation Projects within a Smart City Ecosystem: Cases of Germany and South Korea

    by Olga Shvetsova, Anastasiya Bialevich, Jihee Kim, Mariia Voronina

    This research focused on a behavioral model as a significant tangible enabler for smart city plans and initiatives across Asian and EU regions as per transportation projects. This study aimed to create a behavioral model to serve as a planning tool for policymakers, planners, and implementers of transportation initiatives in smart cities. The paper discusses the validity of the proposed model framework for fostering the diffusion of a successful smart city project transformation in a general smart city ecosystem and particularly within the transportation industry. The framework was verified using three different methods: literature review to give a speculative understanding of current smart city approaches; case studies from Germany and South Korea smart city ecosystems that were selected and applied against the behavioral model; and finally, desktop research (behavioral model) performed for smart city project development. As a result, the authors recognized key variables for deriving a possible successful behavioral model as a suggested efficient framework for further smart city strategic projects. Researchers developed and tested, with two validated examples, the suggested behavioral model for smart city projects with a focus on the transportation industry. Results of this survey could help stakeholders in different countries analyze factors influencing decision-making processes and adopt smart city projects to local business environments.