Generative AI cybersecurity and resilience.

Area of Science:

• Computer Science and Engineering
• Information Science
• Cybersecurity and Ethics

Background:

• Generative Artificial Intelligence (AI) systems enable autonomous content creation across diverse domains, representing a significant advancement in machine learning.
• The rapid evolution and deployment of generative AI present substantial ethical, security, and privacy challenges that current governance frameworks struggle to address.
• Existing socio-technical and governance models require adaptation to effectively manage the implications of advanced AI technologies.

Purpose of the Study:

• To systematically investigate the ethical, security, and privacy challenges posed by generative AI deployment.
• To develop an integrated theoretical framework for evaluating and guiding the responsible application of generative AI across its lifecycle.
• To identify the disconnect between generative AI adoption and institutional safeguard maturity, and propose adaptive governance solutions.

Main Methods:

• A PRISMA-guided systematic literature review was conducted to gather relevant research on generative AI challenges.
• Thematic and quantitative analyses were employed to interrogate the socio-technical implications of generative AI.
• An integrated theoretical framework was developed, drawing on technology adoption, cybersecurity resilience, and normative governance models.

Main Results:

• A significant gap exists between the rapid adoption of generative AI systems and the development of mature institutional safeguards.
• New risks, including those associated with 'shadow Artificial Intelligence,' emerge due to inadequate governance.
• The study identifies a need for adaptive, sector-specific governance strategies to address the unique challenges of generative AI.

Conclusions:

• Responsible deployment of generative AI requires a proactive and adaptive governance approach.
• The proposed five-stage lifecycle framework (design, implementation, monitoring, compliance, feedback) provides a schema for ethical and secure AI application.
• Implementing sector-specific, adaptive governance is crucial for mitigating risks and ensuring the secure application of AI in critical infrastructure.