AGIBOT G2 Humanoid Robots Complete 24-Hour Convenience Store Trial in Shanghai

AGIBOT's G2 wheeled humanoid robots completed a 24-hour trial in a Shanghai convenience store, performing a combined 1,730 tasks, with four of the six robots passing a performance assessment. This pilot demonstrates embodied AI capabilities in a live retail environment, testing adaptability to dynamic conditions.

What happened

The Robot Report states that six AGIBOT G2 robots operated from 10 a.m. on November 2 to 10 a.m. on November 3 in a Shanghai convenience store. During this 24-hour period, the robots executed 1,730 tasks, including 683 customer-service interactions, 369 product pickups and deliveries, 147 coffee orders, 126 breakfast servings, 277 shelf inspections and restocking tasks, and 128 table-cleaning tasks. The trial was livestreamed to over two million viewers.

Four of the six G2 robots passed the final performance assessment, which evaluated time on duty, task completion, customer feedback, response times, and handling unexpected situations. The trial, concluding AGIBOT’s AI WAVE technology showcase, aimed to integrate robotics data, simulation, AI models, hardware, and operational software in a real-world setting. Evaluators conducted practical tests, such as distinguishing between similar products, correcting misplaced items, and navigating around obstacles like people and boxes.

The Robot Report notes that while a convenience store appears simple, tasks like grasping a bag of candy or a slice of cake require precise manipulation, and the environment constantly changes with customer movement and product variations. The G2 robots reportedly used real-time visual information to identify misplaced merchandise, detect shelf gaps for restocking, and distinguish spills from counter patterns for cleaning. AGIBOT's Genie Master embodied AI agent framework was tested for task coordination, error correction, and learning from operational experience. This framework, through its Embodied Execution Protocol (EEP), links task planning, execution, and feedback, enabling robots to improve performance by retaining experience and refining action policies. AGIBOT plans to expand testing across additional roles and environments to improve reliability and adaptability. According to The Robot Report, AGIBOT announced in September 2026 that its 20,000th robot had rolled off the production line.

Why it matters for small companies

This 24-hour trial of AGIBOT’s G2 robots offers a glimpse into the potential for embodied AI in retail and logistics, but it remains a pilot, not a commercial rollout. For small company owners and team leads, the reported 1,730 tasks completed by six robots over 24 hours demonstrates functional capability in a dynamic environment. However, the fact that only four of the six robots passed the performance assessment indicates that these systems are still in active development, not yet mature for widespread, unsupervised deployment.

The ability of the G2 robots to adapt to changing conditions—like customers moving, products shifting, and distinguishing between similar items—is a critical step forward from fixed-automation systems. AGIBOT's focus on learning from operational feedback via the Genie Master framework is crucial for improving uptime and reducing human intervention, which are key metrics for any small business considering automation. However, the article does not provide data on critical factors such as cost per hour of operation versus human labor, specific safety certifications (e.g., ISO 10218, ISO/TS 15066), or mean time between failures (MTBF). Without these metrics, it is difficult to assess the immediate economic viability or regulatory readiness for deployment in a small business setting. The mention of 20,000 robots by September 2026, while ambitious, is a future projection and does not reflect current market availability or verified unit counts.

What to do this week

1. Identify specific, repetitive tasks within your operations that are currently performed manually and research existing, certified robotic solutions (e.g., mobile robots for inventory, collaborative arms for pick-and-place) that are commercially available today.
2. Assess the environmental consistency of potential automation areas in your business; prioritize tasks in predictable, structured environments where current generation automation can be deployed reliably without requiring advanced embodied AI capabilities.
3. Monitor developments in embodied AI and humanoid robotics from companies like AGIBOT, focusing on verified deployment numbers, detailed performance metrics (uptime, payload, cycle time), and explicit safety certifications relevant to human-robot interaction.

Sources

• Six AGIBOT G2 Robots Complete 1,730 Tasks in 24-Hour Shanghai Convenience Store Trial — The Robot Report