Kernel and Mira Launch Industry’s First $300M Secured AI API

GlobeNewswireFebruary 6, 20255 min

KernelDAO, a top restaking protocol with over $2 billion in Total Value Locked (TVL), is pleased to announce its partnership with Mira, a pioneer in trustless verified AI technology, for the launch of a first-of-its-kind API that enhances AI reliability through trustless verification.

The partnership is supported by $300 million in TVL, leveraging KernelDAO’s restaked economic security through one of its key products, Kernel. Through Mira’s verification technology, which achieves 10x higher reliability than traditional models, the joint API is aimed towards ensuring reliability and quality of AI outputs.

Through a dynamic allocation mechanism, the restaked security is distributed across a network of specialized node operators who serve and validate AI model outputs. The economic security is automatically redistributed based on operator performance and demand for specific AI models.

“With this partnership, we are excited to pioneer a new era of AI systems that ensure higher reliability and minimal downtime for developers and businesses. With Kernel’s robust restaking infrastructure and Mira’s advanced AI verification technology, we aim to empower users with AI insights without second-guessing them,” said Amitej Gajjala, Co-Founder of KernelDAO.

The collaboration will focus on addressing key challenges in the current industry including reducing AI system downtime and errors through trustless verification. This is particularly crucial in the present environment with models like DeepSeek, where accuracy metrics are not readily available, and could potentially lead to AI hallucinations and biases.

By integrating Mira’s verification technology, this partnership enables autonomous AI operations with built-in quality assurance, providing developers with verifiable guarantees for AI model outputs, as well as simplifying integration of reliable AI systems into existing applications.

“Until now, developers have faced a critical challenge: how to trust AI outputs in production environments. Our partnership with Kernel introduces real economic consequences for AI verification guarantees, fundamentally changing how developers can deploy AI systems,” said Karan Sirdesai, CEO and Co-Founder of Mira.

With over 400,000 active users, Mira is redefining the AI landscape with an innovative system that combines advanced consensus mechanisms and verification protocols. Combined with Kernel’s premier restaking platform, Mira’s technology will tackle common issues like error rates and complex reasoning, paving the way for truly autonomous AI systems.

The API will be available to developers over the next 12 months, opening doors for new applications and integrations, and allowing them to explore the future of AI in a decentralized world.

The latest partnership comes at the back of KernelDAO’s $40 million Ecosystem Fund that secured support from prominent venture capital firms and leading investors such as Laser Digital, SCB Limited, Hypersphere Ventures, Cypher Capital, ArkStream, and Levitate Labs. With the support of over 20 prominent Dynamic Validation Networks (DVN’s), Kernel is set to expand its ecosystem by adding over 45 strategic partners through the Ecosystem Fund.

GlobeNewswire

newOriginal-white-FinTech1-1

We are one of the world’s leading Fintech-based media publication with our content strategized and synthesized to fit right into the expanding ecosystem of Finance professionals. Be it fintech live news, finance press releases, tech articles from Fintech evangelists or interviews from top leaders from global fintech firms, we give the best slice of knowledge topped up with the aptest trends. Our sole mission is to help tech and finance professionals step up with the rapidly emerging Fintech civilization and gain better insights to emerge victorious in every possible way. We adopt a 360-degree approach in order to cater to present a holistic picture of the fintech arena.

Our Publications



FintecBuzz, 2025 © All Rights Reserved