💻 computer science

Understanding Adoption of AI-Based Cybersecurity Systems: The Roles of Cybersecurity Awareness, Trust, and Digital Literacy

This study utilizes Structural Equation Modeling to demonstrate that while AI-driven cybersecurity awareness and digital literacy significantly promote the adoption of AI-based cybersecurity systems, trust in these systems does not directly influence behavioral intention, highlighting the critical need for organizations to prioritize cybersecurity awareness and digital literacy development over trust-building alone.

Original authors: Abdulkadir Jeilani Mohamud, Abdifatah Nour Rage, Mohamed Adam Isak, Abdisatar Ibrahim Arabow, Abdiwahab Osman Siad

Published 2026-07-08

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Abdulkadir Jeilani Mohamud, Abdifatah Nour Rage, Mohamed Adam Isak, Abdisatar Ibrahim Arabow, Abdiwahab Osman Siad

Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to get a whole neighborhood to install a brand-new, super-smart security system. This system uses Artificial Intelligence (AI) to watch over your house, predict burglars before they arrive, and lock the doors automatically. It's a fantastic tool, but many people are hesitant to let it in.

This research paper is like a detective story trying to figure out why some people are eager to install this smart system while others are scared to touch it. The researchers surveyed professionals in Somalia to understand what goes on in their minds when they think about using AI for cybersecurity.

Here is the breakdown of their findings, using simple analogies:

The Main Characters (The Factors)

Cybersecurity Awareness (The "Flashlight"):
This is how well people understand the dangers of the digital world and how AI can help fight them. The study found that when you shine a bright "flashlight" of knowledge on how these AI systems work, people are much more likely to want to use them. Knowing the enemy (hackers) and the weapon (AI) makes people feel ready to act.
Digital Literacy (The "Driver's License"):
This is the actual skill to use computers and technology. The researchers found this was the most important factor. Think of it this way: You might know that a Ferrari is the best car for racing (Awareness), and you might trust the mechanic who built it (Trust), but if you don't have a driver's license (Digital Literacy), you aren't going to get behind the wheel. The study showed that people who feel confident in their tech skills are the ones most likely to adopt the AI security systems.
Trust (The "Handshake"):
This is whether people believe the AI is honest, safe, and won't make mistakes. The researchers expected that if people trusted the system, they would use it. Surprisingly, this wasn't the case. The study found that even if people shook hands with the AI and said, "I trust you," that trust alone didn't make them want to use the system. They needed the "driver's license" (skills) to go along with the handshake.
Responsible AI Perception (The "Rulebook"):
This is whether people believe the AI follows ethical rules, treats data fairly, and is transparent. The study found that if people believe the AI has a good "rulebook" (it's ethical and fair), they are more likely to trust it. It's like trusting a referee only if you know they follow the rules of the game.
Perceived Risk (The "Fear of the Dark"):
This is the worry that the AI might fail, steal data, or make a mistake. The study confirmed that if people are scared of the risks, they are less likely to use the system. It's like refusing to drive a car because you're afraid the brakes might fail, even if the car is new.
Prior Experience with Breaches (The "Scar"):
This is whether someone has been hacked or had their data stolen before. The study found that people who have been "burned" before are actually more likely to want this AI system. It's like a person who has been bitten by a dog once being very eager to buy a guard dog to protect them next time.

The Big Discovery: How It All Connects

The researchers built a model to see how these factors talk to each other. Here is the story they uncovered:

The Chain Reaction: When you increase Cybersecurity Awareness (the flashlight), it does two things:
1. It boosts Digital Literacy (gives people the driver's license).
2. It boosts Trust (the handshake).
The Missing Link: The study discovered a surprising twist. While Awareness helps build Trust, that Trust does not directly lead to people using the system. You can trust the AI all you want, but if you don't feel skilled enough to use it, you won't.
The Real Hero: Digital Literacy is the bridge. Awareness makes people more skilled, and being skilled is what actually makes them want to use the AI. So, the path is: Awareness → Skills → Adoption.

What This Means for the "Neighborhood"

The paper concludes that if you want organizations or people to start using AI for cybersecurity, you can't just tell them, "It's safe, trust us." That's not enough.

Instead, you need to:

Turn on the Flashlight: Educate them about the threats and how AI works.
Teach Them to Drive: Focus heavily on training and building their digital skills.
Show the Rulebook: Make sure they know the AI is ethical and fair, which helps build trust, even if trust alone isn't the final push.

In short, the paper says: Knowledge creates skills, and skills create action. Trust is nice to have, but without the skills to use the tool, people will keep the AI security system on the shelf.

Technical Summary: Understanding Adoption of AI-Based Cybersecurity Systems

Problem Statement

Despite the rapid deployment of Artificial Intelligence (AI) in cybersecurity to detect, predict, and respond to threats in real-time, there is a significant gap in understanding the behavioral factors driving the adoption of these systems. While existing literature addresses general AI acceptance or specific sectors like fintech and healthcare, limited research integrates cybersecurity awareness, trust, digital literacy, and risk perception into a unified framework specifically for AI-based cybersecurity systems. Furthermore, the mechanisms through which cybersecurity awareness translates into actual behavioral intention—specifically via trust and digital literacy—remain underexplored. This study addresses these gaps by investigating the relative influence of awareness, trust, competency, risk, and prior experience on users' intention to adopt AI-driven cybersecurity technologies.

Methodology

The study employs a quantitative, cross-sectional research design grounded in an integrated theoretical framework combining the Technology Acceptance Model (TAM), Protection Motivation Theory (PMT), Social Cognitive Theory (SCT), Theory of Planned Behavior (TPB), and Trust Theory.

Data Collection: Data were gathered via a survey from 287 valid respondents in Somalia, comprising professionals from education, banking/finance, industry, and government sectors. The sample was selected using purposive sampling to ensure participants had relevant exposure to digital systems.
Analytical Approach: Structural Equation Modeling (SEM) was utilized using IBM SPSS and AMOS. The analysis included:
- Common Method Bias (CMB) Checks: Harman's single-factor test, full collinearity VIF analysis, and Common Latent Factor (CLF) analysis confirmed that CMB was not a significant threat.
- Measurement Model Assessment: Reliability and validity were established through factor loadings (ranging from .621 to .930), Composite Reliability (CR > .70), Average Variance Extracted (AVE > .50), and the Fornell–Larcker criterion for discriminant validity.
- Structural Model Testing: Hypotheses regarding direct effects and mediation were tested using bootstrapping techniques.
Constructs: The model examined seven primary constructs: AI-Driven Cybersecurity Awareness, Responsible AI Perception, Trust in AI-Based Cybersecurity Systems, Digital Literacy, Perceived Risk, Prior Experience with Cybersecurity Breaches, and Behavioral Intention to Use.

Key Results

The structural model demonstrated an acceptable goodness-of-fit, and the hypothesis testing yielded the following findings:

Direct Effects on Behavioral Intention:
- AI-Driven Cybersecurity Awareness: Significantly and positively influences behavioral intention ( $\beta = .199, p < .001$ ).
- Digital Literacy: Emerged as the strongest predictor of behavioral intention ( $\beta = .321, p < .001$ ).
- Perceived Risk: Significantly and negatively influences behavioral intention ( $\beta = -.113, p = .010$ ).
- Prior Experience with Breaches: Significantly and positively influences behavioral intention ( $\beta = .140, p = .036$ ).
- Trust: Contrary to traditional adoption models, trust in AI-based cybersecurity systems did not significantly influence behavioral intention ( $\beta = .051, p = .257$ ).
Antecedent Relationships:
- Awareness to Trust: AI-driven cybersecurity awareness positively influences trust ( $\beta = .206, p < .001$ ).
- Awareness to Digital Literacy: AI-driven cybersecurity awareness positively influences digital literacy ( $\beta = .282, p < .001$ ).
- Responsible AI to Trust: Perceptions of responsible AI (ethics, transparency, accountability) positively influence trust ( $\beta = .149, p < .001$ ).
Mediation Analysis:
- Digital Literacy: Partially mediates the relationship between AI-driven cybersecurity awareness and behavioral intention. The indirect effect was significant ( $\beta = .107, p < .001$ ), indicating that awareness enhances adoption intention both directly and by improving digital competencies.
- Trust: Did not mediate the relationship between awareness and behavioral intention. The indirect effect was insignificant, suggesting that while awareness builds trust, this trust does not directly translate into adoption intention without the presence of digital literacy.

Key Contributions

The paper makes four primary contributions to the literature:

Unified Framework: It integrates awareness, trust, risk, competency, and experience perspectives into a single conceptual model (Trust–Awareness–Competency Framework) specifically for AI-based cybersecurity adoption.
Responsible AI and Trust: It empirically validates the role of responsible AI perception in shaping trust within cybersecurity contexts.
Mediating Mechanisms: It clarifies the indirect pathways of adoption, demonstrating that digital literacy is a critical mediator, whereas trust acts as an outcome of awareness but not a direct driver of intention in this specific context.
Contextual Insight: It provides empirical evidence from an emerging digital environment (Somalia), challenging the assumption that trust is always a direct determinant of AI adoption in security-sensitive domains.

Significance and Implications

The authors posit that the study offers valuable theoretical and practical insights for policymakers, practitioners, and organizations.

Theoretical Significance: The findings extend technology adoption literature by suggesting that in AI-driven cybersecurity, digital literacy is a more potent driver of adoption than trust. The study nuances the role of trust, indicating that while awareness builds trust, adoption ultimately depends on the user's competency to interact with the technology.
Practical Significance:
- Prioritizing Literacy: Organizations should prioritize digital literacy development and AI literacy training, as this is the strongest predictor of adoption intention.
- Awareness Campaigns: Cybersecurity awareness programs are essential not only for threat recognition but also for enhancing digital competencies and building trust.
- Responsible AI Governance: To foster trust, organizations must implement transparent, ethical, and accountable AI governance practices.
- Risk Mitigation: Strategies must address perceived risks (privacy, bias, uncertainty) to prevent them from hindering adoption.

The paper concludes that successful adoption of AI-based cybersecurity systems requires a holistic approach that combines cybersecurity awareness, responsible AI governance, and robust digital literacy development, rather than relying solely on trust-building measures.