Encord Secures €50M to Power Physical AI with Data
▼ Summary
– Encord, a London-based startup, has raised €50 million in a Series C round led by Wellington Management, bringing its total funding to approximately €93 million.
– The investment targets the growing field of physical AI, which powers robots and autonomous systems using complex, real-world data streams.
– The company provides a unified data infrastructure platform to manage the full lifecycle of multimodal data, which legacy tools struggle to handle.
– Encord’s leadership argues that for physical AI, data quality and readiness are the critical bottleneck, not model size.
– The funding reflects a broader investment trend in AI infrastructure, and Encord plans to use the capital to accelerate its product roadmap and market expansion.
London-based startup Encord has successfully raised €50 million in a Series C funding round, signaling strong investor confidence in the infrastructure needed for artificial intelligence that interacts with the physical world. The investment was spearheaded by Wellington Management, with contributions from existing investors like Y Combinator, CRV, and Harpoon Ventures, as well as new supporters Bright Pixel Capital and Isomer Capital. This latest influx of capital brings the company’s total funding to approximately €93 million, providing substantial resources to scale its operations.
The funding arrives as AI technology rapidly expands beyond digital applications into tangible, real-world systems. Encord is building a unified data infrastructure specifically designed for physical AI use cases, which include robotics, autonomous vehicles, and drones. These systems rely on complex, multimodal data streams, such as video, sensor inputs, lidar, and audio, captured in dynamic, unpredictable environments. Traditional data tools, often created for simpler tasks like labeling static images or text, are ill-equipped to manage this scale and variety of information.
Encord’s platform addresses this gap by automating the entire data lifecycle. It handles everything from initial ingestion and organization to detailed annotation, curation, and performance evaluation. This comprehensive approach gives development teams the traceability and insights needed to build, refine, and retrain AI models effectively. The company’s core philosophy is that for physical AI, the primary constraint is not model size but the quality and readiness of the underlying data. Even the most advanced algorithms will fail if trained on inconsistent or poorly curated datasets that don’t reflect actual operating conditions.
The startup reports significant growth, now managing over five petabytes of multimodal data, a threefold increase from just a year ago. Revenue from its physical AI client base has grown by an order of magnitude during the same period. Its global customer list has surpassed 300 teams and includes notable names such as Woven by Toyota, drone delivery leader Zipline, Skydio, and financial services giant AXA.
This investment mirrors a wider trend in AI funding, where venture capital is increasingly flowing into foundational infrastructure, not just model development. Across Europe, substantial rounds are being closed by companies working on everything from AI models and cloud services to autonomy platforms. Encord’s successful raise highlights a belief that the next major wave of AI adoption will occur in embedded systems that operate in the real world, far from traditional data centers.
With this new capital, Encord plans to accelerate its product development and expand into additional markets. The goal is to solidify its software as an essential foundational layer within the growing physical AI technology stack, supporting the reliable and scalable deployment of intelligent machines across various industries.
(Source: The Next Web)
