KnowBe4 Japan Survey Reveals Employees' Security Awareness Regarding AI Agents
KnowBe4 Japan announced the results of its 'Survey on the Security Impact of AI Agents' conducted among employees of Japanese companies and organizations. While there are high expectations for the expanded use of AI agents, concerns about human risks such as information leakage and copyright infringement, stagnation in employees' understanding of risks, and a 'governance gap' have emerged as challenges.
📋 Article Processing Timeline
- 📰 Published: March 31, 2026 at 18:00
- 🔍 Collected: April 1, 2026 at 13:39 (19h 39m after Published)
- 🤖 AI Analyzed: April 21, 2026 at 11:38 (477h 59m after Collected)
KnowBe4 Japan G.K. (Headquarters: Minato-ku, Tokyo; Executive Officer and President: Kazuhiro Riki), a global cybersecurity platform vendor that comprehensively supports human risk and agentic AI risk management, today announced the results of its 'Survey on the Security Impact of AI Agents' conducted among employees working at Japanese companies and organizations.
This survey follows last year's 'Survey on the Security Impact of Generative AI' and was conducted with the aim of understanding the awareness and actual status of security risks associated with the spread of 'AI agents' that autonomously perform tasks.
The survey results revealed that while expectations for the expanded use of AI agents are growing, concerns about 'human risks' such as information leakage and copyright infringement, and the stagnation of employees' understanding of risks, have once again come to light as challenges. In particular, it became clear that in the era of 'agentic AI,' where AI operates autonomously, a 'governance gap' exists, where the definition of risk tolerance and the authority to decide budgets and rules are ambiguous.
In this transitional period where AI agents are deeply permeating business operations, discussions on organizational shared responsibility and the cultivation of a security culture that keeps pace with technological evolution are urgently needed.
## Key Survey Results
* **AI Agent Usage and Future Intent:** 40% of respondents have already introduced AI agents in their organizations, and 70% anticipate expanded use in the future ('strongly agree' 36%, 'somewhat agree' 34%). While last year's generative AI survey showed an adoption rate of 62% and an expectation of expansion of 73%, it appears that the introduction of AI agents is proceeding more cautiously compared to generative AI, as AI agents are more specialized and autonomous tools.
* **Perception and Concerns about Security Risks:** 68% of respondents perceived security risks associated with AI agent use as 'large' ('strongly agree' 27%, 'somewhat agree' 41%). The most frequently cited concerns were 'leakage of confidential information,' followed by 'infringement of copyrights, etc.,' and 'hallucinations (generation of misinformation),' all of which are human risks stemming from human judgment or operation.
* **Current State of Risk Understanding:** Only 30% of respondents stated that 'employees understand' the security risks when using AI agents ('strongly agree' 6%, 'somewhat agree' 24%). This is almost flat compared to last year's generative AI survey (32%), indicating that the 'organizational understanding of risk' continues to lag, regardless of the form of AI.
* **Locus of Security Responsibility:** The security responsibility when using AI agents, whether it lies with management, IT/security departments, or AI agents...
This survey follows last year's 'Survey on the Security Impact of Generative AI' and was conducted with the aim of understanding the awareness and actual status of security risks associated with the spread of 'AI agents' that autonomously perform tasks.
The survey results revealed that while expectations for the expanded use of AI agents are growing, concerns about 'human risks' such as information leakage and copyright infringement, and the stagnation of employees' understanding of risks, have once again come to light as challenges. In particular, it became clear that in the era of 'agentic AI,' where AI operates autonomously, a 'governance gap' exists, where the definition of risk tolerance and the authority to decide budgets and rules are ambiguous.
In this transitional period where AI agents are deeply permeating business operations, discussions on organizational shared responsibility and the cultivation of a security culture that keeps pace with technological evolution are urgently needed.
## Key Survey Results
* **AI Agent Usage and Future Intent:** 40% of respondents have already introduced AI agents in their organizations, and 70% anticipate expanded use in the future ('strongly agree' 36%, 'somewhat agree' 34%). While last year's generative AI survey showed an adoption rate of 62% and an expectation of expansion of 73%, it appears that the introduction of AI agents is proceeding more cautiously compared to generative AI, as AI agents are more specialized and autonomous tools.
* **Perception and Concerns about Security Risks:** 68% of respondents perceived security risks associated with AI agent use as 'large' ('strongly agree' 27%, 'somewhat agree' 41%). The most frequently cited concerns were 'leakage of confidential information,' followed by 'infringement of copyrights, etc.,' and 'hallucinations (generation of misinformation),' all of which are human risks stemming from human judgment or operation.
* **Current State of Risk Understanding:** Only 30% of respondents stated that 'employees understand' the security risks when using AI agents ('strongly agree' 6%, 'somewhat agree' 24%). This is almost flat compared to last year's generative AI survey (32%), indicating that the 'organizational understanding of risk' continues to lag, regardless of the form of AI.
* **Locus of Security Responsibility:** The security responsibility when using AI agents, whether it lies with management, IT/security departments, or AI agents...