a16z Crypto invests $15 million to support the Decentralization data infrastructure project Poseidon

The decentralized AI data infrastructure project Poseidon has recently completed a $15 million seed round of financing, led by a16z crypto. This funding will be used to accelerate the construction of a full-stack data infrastructure layer aimed at robotics, multimodal models, and next-generation physical AI, with the goal of addressing key bottlenecks in AI development: high-quality, intellectual property-secure, and traceable data acquisition.

Sandeep Chinchali, Chief AI Officer of Story and Chief Scientist of Poseidon, stated: "In today's world where models and computing power are gradually being commoditized, the true competitive barrier is data."

Poseidon's goal is to extend the programmable intellectual property layer built by Story into real-world data and AI training scenarios. Story is committed to creating an underlying protocol that supports programmable ownership, licensing, and attribution of creative and AI assets, while Poseidon further applies these functionalities to support data circulation and compliant use for AI.

AI moves from digital to physical, data bottlenecks intensify

In the past two years, AI has made significant breakthroughs in the fields of text and images. However, as AI expands into physical world scenarios such as robotics, autonomous driving, and smart devices, the demand for real-world data has surged.

However, the existing data collection model faces three major structural challenges:

1. Scarcity of long-tail data: The collection cost of high-value data (such as POV videos, 3D simulations, edge sensors) is high and the sources are dispersed.
2. Licensing and compliance risks: The traditional data chain is complex and copyright is unclear, making it difficult for enterprises to safely access.
3. Lack of incentive mechanisms: Data contributors lack ownership protection and profit distribution, leading to insufficient willingness to participate.

The next generation of AI models not only needs to "understand the world", but also to "comprehend and act upon the world".

Poseidon: The Data Operating System Built for AI

To address the above challenges, Poseidon provides a comprehensive operational system covering collection, labeling, licensing, circulation, traceability, and incentives, transforming data from "resources" into "assets" that have legal validity and economic incentive mechanisms.

S.Y. Lee, CEO and co-founder of Story and President of Poseidon, stated, "With Story's programmable intellectual property layer and immutable intellectual property registration system, we ensure that each dataset is licensed, traceable, and enforceable. Coupled with our integrated licensing module, the team can seamlessly complete data authorization, monetization, and usage without worrying about legal risks. This is not just infrastructure development, but also empowering AI systems that can be truly deployed in the real world."

Poseidon is a data operating system designed for AI training processes, with native intellectual property tracking and programmable incentive capabilities at every stage, from collection to labeling, from licensing to circulation. Its core functions include:

• Full-stack data pipeline: Supports standardized collection and cleaning of multi-source data (POV videos, sensors, voice, synthetic simulations)
• On-chain intellectual property registration and traceability: All data is recorded as intellectual property assets on the Story Protocol, ensuring clear origin and legal authorization.
• Incentive and Permission Module: Contributors, Annotators, and Synthetic Models can earn profit distribution through smart contracts to achieve sustainable growth of the data network.
• Composable data marketplace: Developers can filter, call, and integrate data by license, automatically triggering permissions and revenue sharing, reducing copyright risks.

Poseidon provides a truly compliant, composable, and sustainable data network for AI model training, becoming an important infrastructure for AI to shift from model-driven to data-driven.

Building the Data Economy Foundation in the AI Era

Chris Dixon, founder and managing partner of a16z crypto, stated: "AI foundational models have exhausted the easiest-to-access training data. Poseidon's Decentralization data layer is attempting to establish a new economic foundation for the internet, rewarding creators and data providers who provide diverse inputs for the next generation of intelligent systems. We are excited to support Poseidon in addressing one of the most critical bottlenecks in AI development."

Poseidon has partnered with several leading AI companies and will use this round of financing to expand its tech stack, including an SDK toolkit, a data contributor workbench, and a licensing and revenue-sharing management module. The early access channel will open this summer, allowing registration for AI developers and data contributors.

About Poseidon

Poseidon is a decentralized full-stack data layer tailored for AI training processes. Incubated by the Story Protocol team, it aims to connect high-quality data supply with AI model demand. The platform supports the full-link process from data contribution and licensing to verification and integration, ensuring that data possesses security, compliance, and economic incentive mechanisms.