An experiment at LBX demonstrates the use of a humanoid robot in retail for repetitive product-finding and picking operations. The available description does not confirm the pharmacy context, retrieval of an order from a pharmacy system, or transfer of the picked order to packing, so these details cannot be considered established.
The confirmed material supports only a more general scenario: the robot participates in finding and picking products in a retail environment. The available data do not show exactly how the system links a digital task to a physical item, chooses a route, or records the result.
What a pharmacy scenario from order to packing might look like
The detailed sequence of G1's operation is not described in the available material. It has not been confirmed that the scenario begins with an order entering the pharmacy system, after which the robot receives a task, finds a shelf, recognizes the package, and moves the product to the packing area.
To verify such a scenario, it is necessary to establish separately whether the system performs the following functions:
- receiving a task from the order management system;
- navigating the pharmacy space;
- finding the required shelf and specific item;
- visually recognizing the package;
- grasping the product from the shelf;
- handing the picked order to an employee or an automated packing area.
The experiment confirms the technology's area of application but does not disclose how order contents are checked. For a pharmacy scenario, this issue would need to be assessed separately: the available materials do not show what data the system checks, how it records each item, or what happens when there is a discrepancy.
Which tasks in a pharmacy scenario are suitable for automation
Based on the available evidence, it is possible to speak of using the robot for repetitive product-finding and picking operations. It cannot be claimed that G1 has already demonstrated suitability for a specific set of pharmacy tasks or that its deployment is particularly logical under certain storage and inventory conditions.
A comparison of the pharmacy scenario with a warehouse, as well as claims about similar packages, small print, reflective film, expiration dates, and dispensing procedures, requires separate confirmation. The provided material contains no such data.
A checkpoint before packing can be considered a possible way to organize the process, but the available material does not confirm that this is how the LBX experiment is structured. Information is needed about who checks the picked items and how the result is recorded.
What performance metrics GBOT claims
In the available transcript, GBOT claims that one G1 can process from 400 to 500 orders per day. The company also states an average time of about 68 seconds per order and less than 90 seconds before the products are handed over for packing. These figures are identified as company claims and have no independent confirmation in the provided materials.
| Metric | GBOT claim | How to interpret it |
|---|---|---|
| Daily throughput | 400–500 orders | The throughput of one robot claimed by the company |
| Average time per order | About 68 seconds | The claimed average; measurement conditions were not disclosed |
| Time until handover for packing | Less than 90 seconds | The interval claimed by the company; the stages included are not described |
| Picking accuracy | 100 percent | A manufacturer's claim requiring independent verification |
| Recognizable assortment | More than 10 000 products | The claimed number of items; the counting methodology was not provided |
These figures cannot be considered an industry standard or a confirmed result for any pharmacy. The available material does not indicate the assortment, number of items per order, route, or amount of manual intervention under which they were obtained.
Comparing the results with another pharmacy would require an identical methodology. The order composition, method of calculating time, and rules for recording stops and repeated grasps must be defined in advance, but the testing conditions themselves are not specified in the provided data.
Accuracy and assortment: where the manufacturer's claim ends
GBOT claims one-hundred-percent picking accuracy. The company also reports that G1 recognizes and picks more than 10 000 different products, including prescription and over-the-counter medicines, medical devices, and other pharmacy products. In the available transcript, this information is presented as company statements rather than the results of independent verification.
A claim about recognition does not establish the correctness of the entire pharmacy operation. The provided materials do not show how the system distinguishes dosage, dosage form, batch, or expiration date, or verifies that an item matches a specific order.
What happens if the picking record does not match the contents of the package? The available data do not answer whether task history, images, scans, and an employee action log are maintained. Without a description of these mechanisms, it is impossible to assess how an operator would identify the stage at which the discrepancy arose.
Prescription products require separate controls
GBOT's claim that it works with prescription medicines does not by itself confirm that G1 makes dispensing decisions about them. The provided materials contain no description of prescription verification, authorization to dispense, restrictions, or final handover.
Before deploying such a scenario, it would be necessary to define access controls, task confirmation, item-by-item verification, and the procedure for stopping when the system is uncertain. The available information does not show whether these functions are implemented in G1 or who is responsible for verification.
The same applies to product substitution. The available description contains no data on whether the robot offers an alternative item or whether confirmation of such an operation is required.
Why the pharmacy scenario differs from warehouse automation
The available material confirms an experiment in retail but does not describe the LBX environment sufficiently to draw conclusions about its similarities to or differences from a warehouse or pharmacy. The characteristics of the aisles, product displays, presence of customers, and robot movement near people have not been confirmed.
The claim that the humanoid design offers advantages also requires separate verification. The provided data do not show that G1 uses shelves at human height or makes it possible to avoid restructuring the sales floor for conveyors and specialized storage cells.
The LBX experiment demonstrates the use of humanoid robots in retail for repetitive product-finding and picking operations. It does not prove that G1 is ready to replace pharmacy workers or perform the entire order-processing workflow.
How to assess a robot's readiness for deployment
- Describe the process. Record the order's path from receipt to packing and distinguish the robot's confirmed functions from those that are only assumed.
- Separate the assortment. Determine which product characteristics and storage conditions need to be taken into account during testing.
- Define stopping points. Establish in advance the procedure to follow when a product is unavailable, recognition is uncertain, or the data do not match.
- Check integration. Clarify which systems are connected to the robot and how the order, storage location, and picking result are recorded.
- Conduct an independent measurement. Speed and accuracy must be calculated using described orders, including errors, repeated grasps, stops, and manual interventions.
- Assess the economics. The calculation includes the robot, its maintenance, integration, staff training, and a backup process in case of downtime, if these items are included in the project.
Which retail operations could come next
After the picking results have been confirmed, other repetitive operations involving finding, moving, and returning products can be considered for further assessment. The provided materials do not confirm that G1 already restocks shelves, moves boxes between areas, or processes returns.
The available description says nothing about advising customers, selecting therapy, assessing complaints, or substituting medicines. Such functions cannot be attributed to the robot, nor can their potential for automation be inferred from a single experiment.
Conclusion
The LBX experiment demonstrates the use of a humanoid robot in retail for finding and picking products. The available material does not confirm the complete route from a pharmacy order to packing.
GBOT claims a throughput of 400–500 orders per day, a time of about 68 seconds per order, handover for packing in less than 90 seconds, one-hundred-percent accuracy, and operation with more than 10 000 products. These figures should be treated as manufacturer's claims because the provided materials contain neither independent verification nor a measurement methodology.
The main question for pharmacies is which functions G1 actually performs and how the system records the result. Without a description of integration, item verification, and handling of exceptional cases, it is impossible to determine whether the robot will become a working component of pharmacy logistics or remain a demonstration solution.
FAQ about the G1 robot and the pharmacy scenario
What does G1 do in a pharmacy scenario?
The available material confirms the robot's participation in an LBX experiment involving the finding and picking of products in retail. Receiving an order, finding products on shelves, and handing them over for packing as a complete process have not been confirmed.
How many orders per day does GBOT claim?
The company gives a range of 400 to 500 orders per day for one robot. The provided materials contain no independent confirmation of this figure.
Has one-hundred-percent picking accuracy been confirmed?
No. This is a claim by GBOT that requires verification in an independent test with specified conditions and a defined counting methodology.
Can G1 dispense prescription medicines independently?
There is no confirmation that G1 independently makes decisions about dispensing prescription medicines. The available description supports only the company's claim that it recognizes and picks more than 10 000 products, including prescription and over-the-counter medicines.
Compare models before you start
The service sets its plans, limits and model catalog. If they differ from this article, contact us so we can update it and record a new review date.
Browse modelsAffiliate link: your price stays the same and the project earns a commission.
