日本語 
SentinelOne Japan Co., Ltd., the Japanese subsidiary of SentinelOne®, a global leader in autonomous cybersecurity, announced on April 21, 2026, that it has entered into a strategic partnership with identity security company Silverfort to protect human, AI agent, and other non-human identities (NHI).
This partnership integrates industry-leading runtime security across identity, endpoints, cloud workloads, and AI applications into a single platform, creating a robust security posture to protect organizations from increasingly sophisticated identity attacks. As a result, organizations can confidently leverage agent-based innovations while establishing an autonomous defense system that can detect and respond to both human and agent threats at machine speed.
As enterprises accelerate the adoption of AI-driven systems and agent-based platforms, new forms of identity risk are emerging. The modern enterprise environment is comprised of diverse “workers”—service accounts, APIs, workload identities, and increasingly autonomous AI agents. These AI agents perform actions at machine speed and scale, replacing humans and systems. Remarkable cyberattack incidents in recent weeks vividly illustrate the realities security professionals will face. The challenge lies in how quickly organizations can respond to agent-based and identity-based threats in today’s increasingly automated enterprise environment.
Also Read: HCLTech and CrowdStrike Strengthen Cyber Risk AI
On March 31, 2026, a state-sponsored threat actor in North Korea hijacked the npm credentials of a key Axios maintainer and executed a sophisticated supply chain attack. The attack detected its first infection just 89 seconds after its public release, a speed that would have been impossible to defend against with manual workflows.
A week prior, SentinelOne detected and blocked a Trojanized version of LiteLLM that had been tampered with by an autonomous coding assistant . It identified and preemptively stopped a malicious process chain being executed with unlimited privileges by Anthropic’s Claude Code. This malicious activity was carried out by a normal automated workflow without human intervention. In this case, SentinelOne’s behavioral AI detected the Trojanized package mid-execution and preemptively stopped it within 44 seconds.
These two cases are examples of attacks on third-party software supply chains, but both demonstrate how quickly trusted authentication methods can be exploited and how quickly security teams must respond.
Through this partnership, Silverfort and SentinelOne will directly address this challenge by protecting identities at runtime, enabling faster containment and mitigating lateral movement and privilege escalation. Combining Silverfort’s expertise in discovering and protecting AI agents and non-human identities with SentinelOne’s leadership in AI-driven detection and deployment to AI security will establish a security foundation that protects environments where humans, machines, and AI agents operate simultaneously and autonomously. The impact on customers will be tangible. Identity risk has become a critical signal in AI-driven threat detection and automated SOC (Security Operations Center) workflows, enabling the blocking of fraudulent authentication requests at runtime and the isolation and containment of compromised credentials.
This partnership goes beyond mere integration, evolving into deep technological collaboration and joint research, directly integrating identity enforcement into autonomous security operations. By closely linking Silverfort‘s runtime identity security with SentinelOne‘s AI-powered Singularity™ security, the two companies will build a real-time, unified control plane that integrates identity and endpoint intelligence into a single decision fabric, enabling real-time enforcement across both endpoint and identity domains.
SOURCE: PRTimes
English