Dream Labs Unveils Dream-1 API Turning Robot Skills Into Video Prediction Tasks
The startup provides a universal foundation model that treats robotic skills as video prediction tasks, charging $3 per hour for access.
Key points
- Dream Labs secured $90 million in seed funding led by B Capital and Kindred Ventures.
- The new Dream-1 model controls unseen bimanual robots on novel tasks using video prediction.
- The public API is live, aiming to reduce the months-long programming process for industrial automation.
Dream Labs has launched a public application programming interface for its Dream-1 robot foundation model, coinciding with a $90 million seed funding round. According to SiliconANGLE, the startup aims to address the limited intelligence of current industrial robots by providing a universal control system. The round was jointly led by B Capital and Kindred Ventures, with participation from A.Capital, Unusual Ventures, Wing VC, SV Angel, Scribble VC and angel investor Adrian Aoun.
Root cause
The company identifies a lack of expertise and knowledge as the primary barrier to more capable robotics. While hardware costs have fallen, with dextrous arms selling for thousands of dollars, these devices ship without inherent intelligence. Buyers must manually teach each arm specific jobs through tedious, specialized programming. This process can take months and must be repeated from scratch whenever new hardware is introduced, preventing knowledge transfer between machines.
Attack path
This section is not applicable as the source material describes a commercial product launch and funding event, not a security vulnerability or cyber incident. There is no attack vector, exploit method or threat actor activity described in the provided text.
Affected versions
The source material does not list software versions or affected products in the context of a vulnerability. Dream Labs is releasing Dream-1, its first public foundation model. The text notes that over five million industrial robots currently operate globally, most designed for single tasks. These existing systems are the target market for the new API, which seeks to augment their capabilities without requiring full reprogramming.
Mitigation
This section is not applicable as the source material does not describe a security flaw, breach or risk requiring mitigation. The text focuses on the technical approach of Dream-1, which treats robotic skills as simple video prediction problems, allowing it to control robots it has never seen in new environments.
What it means for organisations
Teams usually evaluate new automation suppliers for long-term viability and integration complexity. Check whether the vendor’s model aligns with existing hardware standards and data privacy requirements. Verify that the API documentation covers error handling and latency expectations for real-time control. Ensure the procurement and engineering leads are informed about the shift from single-task programming to foundation model reliance. Assess how this change impacts internal training workflows and whether current staff possess the skills to manage model-based automation rather than direct code execution.
Background: APIs
APIs allow your application to request data from a server on demand. Webhooks allow a server to send data to your application when specific events occur. APIs pull data; webhooks push it. Choose APIs for user-initiated actions and webhooks for asynchronous event notifications.
Read the full guide: APIs vs Webhooks: Choosing the Right Integration Pattern
What to do and how to stay safe: Dream Labs
- Review current robotic hardware contracts to identify clauses related to software updates and third-party API integration.
- Assess data governance policies to ensure video feeds used for model inference comply with internal security and privacy standards.
- Monitor the vendor’s documentation for changes in pricing, rate limits or model versioning that could affect operational continuity.
- Establish a sandbox environment to test the new API’s latency and reliability before deploying it in critical production workflows.
General security guidance from the TechSec Observer newsroom. It is not confirmed advice from the organisations named in this story.
Frequently asked questions
How much did Dream Labs raise in its latest funding round?
Dream Labs raised $90 million in seed funding, according to SiliconANGLE.
What is the cost of using the Dream-1 API?
The API costs $3 per hour for access to the universal robot foundation model.
How does Dream-1 train robots for new tasks?
The model treats robotic skills as simple video prediction problems, allowing it to control robots on tasks it was never shown.
