Tokyo, Japan - June 23, 2025- Priv Tech, Inc., a leading privacy and cybersecurity company based in Tokyo, is pleased to announce that Adrian Oprea, a globally recognized ethical hacker (handle name: jungletsubasa), has joined the company's security teams. His expertise will significantly enhance Priv Tech's capabilities in addressing advanced cybersecurity threats and emerging AI-related vulnerabilities. His addition also marks the official launch of Priv Tech's new LLM Application Security Assessment service.
With cyberattacks becoming increasingly sophisticated and the rapid adoption of generative AI introducing novel security risks, organizations are facing a new class of challenges. In response, Priv Tech is strengthening its technical capabilities by welcoming Adrian Oprea--an internationally recognized expert in vulnerability discovery--whose experience brings practical and impactful enhancements to our security services.
About Adrian Oprea
Adrian Oprea has resided in Japan since 2010 and has worked in the cybersecurity field since 2016. He began as a vulnerability assessor and became an active ethical hacker in 2019. Oprea is known for identifying and responsibly disclosing numerous vulnerabilities across both domestic and international systems. Notably, he is credited with reporting CVE-2024-37031, a vulnerability officially registered in the global CVE database.
In April 2025, he placed 15th out of 4,000 participants in the Gray Swan Arena, a prestigious global security competition organized by Gray Swan AI. He is also involved in top-tier bug bounty platforms such as Synack and HackerOne, where he has earned high rankings--including a first-place finish on a HackerOne project--underscoring his reputation as a trusted ethical hacker in the global security community.
LLM Application Security Assessment Now Available
With Oprea's expertise, Priv Tech has officially launched its LLM (Large Language Model) Application Security Assessment service. This new offering is designed to identify and mitigate unique security threats in generative AI applications through a hybrid approach that combines advanced tools and manual testing by expert ethical hackers.
Key areas of focus include:
- Prompt Injection Attacks
・Direct Prompt Injection
・Indirect Prompt Injection
- Jailbreak Testing
Exploiting LLMs to generate unauthorized or harmful content through prompt-based attacks.
- Bias and Hallucination Risk Analysis
Assessing potential misinformation caused by biased training data or inference errors.
- API Testing
Evaluating the security importance of backend APIs and infrastructure in LLM-based services, and identifying risks such as large-scale misuse due to API key leakage or rate-limit circumvention.
- LLM Agent Risk Assessment
Investigating vulnerabilities in systems using LLM-based agents capable of performing autonomous actions.
About Priv Tech
Priv Tech, Inc. is a pioneering Japanese startup specializing in privacy support and cybersecurity solutions. As one of Japan's top three providers of privacy data solutions, Priv Tech integrates advanced risk mitigation strategies to protect businesses from both regulatory and cybersecurity threats. With an elite team of bilingual security experts, Priv Tech is committed to helping Japanese companies navigate the complex digital landscape with confidence. Our innovative approach and strong customer base position us as a trusted leader in the Japanese market.