The Evolution of Coding in the Digital Transformation Era Cybersecurity Implications of Artificial Intelligence and Low-Code Development

The accelerating wave of digital transformation has made software the core infrastructure of modern competitiveness. Yet, as organizations adopt artificial intelligence (AI) and low-code/no-code tools to accelerate development, they also expose new vectors of vulnerability, opacity and technical debt. This thesis examines how the evolution of software creation, from traditional programming to AI-assisted and low-code paradigms, reshapes the cybersecurity and governance landscape. The research addresses three key questions: (1) how AI-assisted and low-code development alter software vulnerabilities and accountability; (2) which governance and assurance mechanisms can mitigate emerging risks without undermining agility; and (3) how standardized frameworks can institutionalize quality and security in increasingly automated environments. Methodologically, the study combines literature review, risk analysis, and framework design, supported by the Italian initiative Italy New Code as a reference for certification and secure-by-design principles. Findings show that AI-generated and low-code software enhance productivity and democratize innovation but also erode transparency, expand attack surfaces, and increase reliance on opaque models. Addressing these issues requires integrating technical safeguards, automated testing, software bills of materials, provenance tracking, with governance measures such as certification, standardization, and regulatory alignment under NIS2 and the EU AI Act. The thesis contributes a conceptual and operational framework for secure, compliant, and sustainable software development in the AI era, concluding that the future of coding lies not in replacing human expertise but in orchestrating intelligent tools and certified components within transparent, auditable, and collaborative ecosystems where speed and security coexist in balance.