AIM Intelligence to Serve Security Layer for Medical AI Agents
Cariva Thailand Company Inc. and AIM Intelligence Co., Ltd. will be jointly serve the combined solution for deploying safe medical AI models.
Nov 28, 2024
Cariva Thailand Company Inc. and AIM Intelligence Co., Ltd. will be jointly serve the combined solution for deploying safe medical AI models.
In the face of expanding use cases of AI Agents at the medical domain with constraints of safety and security, organizations need help making it safe and aligned. AIM Intelligence provides AI models for alignment, while the Cariva platform provides AI Agents that assist user with medical expertise with integrated workflow process automation. The combined offering empowers organizations to easily deploy and productize Medical AI Agent services.
AIM Intelligence builds on its powerful alignment foundation models and tools, including AIM Red for automated penetration testing and AIM Guard for safety guard-railing. The system can scale evaluation dataset for AI stress testing so developers can easily explore vulnerabilities of target AI model. Users of the combined system can deploy the safe medical AI Agent with safety guardrail that automatically detect and moderate behaviors of AI Agents.
Cariva President, shared, "For medical organizations to unlock the potential of the AI Agent, security must be handled. Our need to instantly find important vulnerabilities of our services for safety and alignment, powerful AIM Red will make our Agents more safe and trustworthy."
Sangyoon Yu, founder and CEO of AIM Intelligence shared, "Manually stress testing is time-consuming and hard to scale. Moreover, misalignment of AI Agents can lead to misinformation or even service failures. With the power of AIM Intelligence’s models available in Cariva, organizations are better equipped than ever before to make the AI safe and trustworthy."
From the web interface, Cariva users can browse the AI safety score, stress test the target model, and monitor them with visualized dashboard. Non-technical users can scale test dataset, red-team, and safeguard their model.
Cariva can manage the safety and security of medical AI model, in any number of cloud and local systems. After review from Cariva, evaluated results can be shared as dashboard with tailored report for executives.
About Cariva
The CARIVA is dedicated to making high value health-related data more accessible yet secured to individuals, healthcare professionals, researchers, and policy makers. They provide data-driven insights to accelerate life-changing innovations in the hopes of better health outcome for all. The Cariva has been in continuous development since 2021 with a rich tool set and scores of integrated technology partners. Cariva is headquartered in the Bangkok with channel partners worldwide.
About AIM Intelligence
AIM Intelligence serves security layer for AI Agents. AIM Intelligence' state-of-the-art AI alignment technology enables the safe and trustworthy deploy of vertical AI Agents that behave within developer’s design and intention. Leading venture capitalists, technology companies, AI luminaries, and successful founders back the company. AIM Intelligence is headquartered in APAC office in Seoul. Learn more at https://aim-intelligence.com/
Share article