Beyond Algorithms: Generative AI as a Catalyst for Accessible and Accountable Optimization

Generative artificial intelligence (GenAI) is reshaping decision-making, yet mathematical optimization, despite its value, remains limited to expert use. This study examines how GenAI can broaden optimization’s reach by reducing technical barriers and fostering wider participation in decision science. Using a descriptive qualitative approach, the analysis integrates Scopus-indexed literature (2019–2025), institutional policy frameworks, and practitioner-oriented sources, including La Academic.

Three perspectives guide the study: the 4I framework (Insight, Interpretability, Interactivity, Improvisation), the Technology Acceptance Model (TAM), and Institutional Theory. Findings show that GenAI strengthens Insight by synthesizing diverse data into summaries, enhances Interpretability by explaining trade-offs in plain language, enables Interactivity through scenario testing, and supports Improvisation during disruptions. These functions reinforce TAM’s constructs of perceived ease of use and usefulness, making optimization more accessible.

Institutional Theory highlights adoption drivers: coercive regulations demanding explainability, normative professional standards, and mimetic pressures from peer adoption. Yet challenges persist, including hallucinations, bias, and over-reliance, which threaten trust and reliability.

The study concludes that GenAI complements rather than replaces optimization, offering accessibility while optimization provides rigor. Responsible adoption requires verification systems, educational reforms, and institutional safeguards, paralleling democratization efforts in academic publishing such as La Academic’s Summary Publishing in Sinta-Accredited Journals (Academic, 2025).