English 
SherLOCK Inc. has developed a prototype of “Agentic Security Hub,” a security solution that will be at the core of next-generation AI infrastructure. This solution enables safe collaboration between autonomously operating AI agents (Agentic AI), and aims to become an essential fundamental technology in the AI agent market, which is expected to expand rapidly in the future.
In recent years, the use of “Agentic AI,” in which AI agents autonomously perform tasks and multiple agents collaborate to solve complex problems, has been rapidly increasing. This innovative technology is expected to accelerate the automation of business processes and the sophistication of decision-making.
However, at the same time, new security issues have emerged as AI agents autonomously access external systems and data. In particular, risks that are difficult to address with conventional AI security solutions, such as prompt injection, access permissions, appropriate use of external tools, supply chain vulnerabilities, and memory poisoning, are becoming a bottleneck in the development of Agentic AI.
こちらもお読みください： UPWARD Launches AI-Powered Voice Analysis Tool to Streamline Field Sales Reporting
Therefore, we have begun developing a new security solution for Agentic AI that combines the flexibility of AI utilization with security governance.
ソース PRタイムズ
日本語