Insider Brief
- AIM Intelligence raised $1.3M in a Pre-A round led by Mirae Asset Capital to advance safety, alignment, and controllability infrastructure for next-generation generative AI systems.
- Its product suite — AIM Red, AIM Guard, and AIM Supervisor — addresses risks like hallucinations, data leakage, and adversarial misuse, with deployments across high-stakes sectors including finance and telecom, and partnerships with organizations such as Anthropic, Meta, and LG Uplus.
- Funding will support expanded research, international growth, and development of multimodal and embodied AI systems, as AIM Intelligence works to set the global standard for generative AI security.
PRESS RELEASE — AIM Intelligence has raised $1.3 million in a Pre-A round as demand for safe, aligned AI reaches unprecedented levels. The round was led by Mirae Asset Capital, with backing from Bluepoint Partners (via the LG U+ Strategic Fund), Z Venture Capital (ZVC), and follow-on investor Mashup Ventures.
This funding will accelerate development of infrastructure to ensure the safety, alignment, and controllability of next-generation generative AI systems.
The company is focused on addressing foundational risks like hallucinations, data leakage, and adversarial misuse as AI systems assume greater decision-making autonomy.
AIM Intelligence’s core product suite includes AIM Red, which automates penetration testing and vulnerability discovery through AI red-teaming; AIM Guard, which provides real-time filtering of unsafe inputs and outputs; and AIM Supervisor, which monitors AI agent reasoning paths to detect policy drift and unsafe cognition.
These technologies have already been adopted across high-stakes sectors including finance and telecommunications. Clients such as KB Kookmin Card, KB Securities, Woori Bank, and LG Uplus have deployed AIM’s tools, with LG Uplus also conducting a pilot project as part of its strategic investment. Further collaborations include generative AI reliability testing with KT, LG AI Research, the Korea Credit Information Service, and the Telecommunications Technology Association (TTA).
AIM Intelligence’s impact extends beyond Korea. In 2024, the company won Meta’s Llama Impact Innovation Award — the only Korean recipient — and was chosen by Anthropic for its private bug bounty program. Through this partnership, AIM Intelligence successfully identified and reported security vulnerabilities in non-public Claude models, including policy bypasses and adversarial prompt vectors.
Its research leadership has been recognized globally, with four papers accepted at top-tier AI conferences including ICML and ACL.
“Our mission is to prevent AI failures before they occur,” said David Yu, CEO of AIM Intelligence. “We are laying the groundwork for Sovereign AI — systems that are institutionally aligned, independently operable, and fundamentally safe.”
With this new fund, AIM Intelligence will deepen its research capabilities, expand internationally , and push forward into new frontiers such as multimodal and embodied AI systems, as it works to establish the global benchmark for generative AI security.
You can find more information at its company website.
Skye Kim
AIM Intelligence
[email protected]
