Sanctuary AI achieves 99.5% success rate on industrial wire-insertion task using existing robot hardware
By
Korthos
Summary
Sanctuary AI, a Vancouver-based robotics company, achieved a 99.5%+ success rate on a plug-insertion task at 2.54-second cycle time for an unnamed global Tier 1 automotive supplier. The system manipulated a flexible wire on a moving conveyor and matched the throughput of the customer's live production line. Notably, the task was performed on two industrial robotic arms rather than Sanctuary's own Phoenix humanoid platform, suggesting a strategic pivot toward applying their Physical AI software to existing industrial hardware rather than waiting for humanoid-specific deployments.
Source

Twitter / XSanctuary AI achieves 99.5% success rate on industrial wire-insertion task using existing robot hardwarekorthos.xyzKey quotes
· 3 pulledSanctuary AI, the Vancouver-based robotics company founded in 2018, says its Physical AI software achieved a 99.5%+ success rate at a 2.54-second cycle time on a plug-insertion task for a global Tier 1 automotive supplier
The system matched the throughput of the customer's live production line although the customer has not been named.
Sanctuary's release shows the task running on two industrial robotic arms, not on Phoenix, its own humanoid platform.
You might also wanna read
AgiBot humanoid robots complete 8-hour tablet inspection shift with 99.9% accuracy, raising cybersecurity questions for industrial AI
The article reports on AgiBot deploying four humanoid robots for an eight-hour live-streamed shift on a tablet assembly line in Nanchang, Ch
undercodetesting.com·3d agoNew Benchmark Reveals High Rates of Outcome-Driven Constraint Violations in Autonomous AI Agents
Researchers introduce a new benchmark for evaluating autonomous AI agents' safety, specifically focusing on outcome-driven constraint violat
NVIDIA Launches Halos Safety System for AI-Powered Robots in Factories and Warehouses
NVIDIA announced Halos for Robotics, a full-stack safety system for "physical AI" — machines that can sense, decide, and act in the real wor
A Field Guide to Production-Ready AI Agents: Context Windows, Security, and Drift Monitoring
Karl Mehta presents a field guide for building production-ready AI agents, focusing on four key engineering challenges: context-window disci
Nvidia's AI coding agents autonomously train robots to install GPUs and cut zip ties
Nvidia has developed a new agent harness framework that enables AI coding agents to autonomously train robotic arms to perform physical task
arstechnica.com·9d agoNVIDIA's AI coding agents autonomously train robots to install GPUs and cut zip-ties
NVIDIA has developed a new agent harness framework that enables AI coding agents to autonomously train robotic arms to perform physical task
arstechnica.com·13d ago
Comments
Sign in to join the conversation.
No comments yet. Be the first.