RLWRLD unveils logistics demo of ‘RLDX-1’ robot foundation model at KDDI SUMMIT 2026

RLWRLD Inc. (RLWRLD) exhibited at KDDI SUMMIT 2026 and presented a live demonstration using its ‘RLDX-1’ robot foundation model. The demo simulated shipping work in a logistics warehouse. The supplied portion of the announcement does not include the exhibition date or figures such as task completion rate or throughput.

RLWRLD demonstrated its ‘RLDX-1’ robot foundation model in a scenario simulating shipping work in a logistics warehouse. The demonstration linked the model to a specific work setting, but the supplied portion of the announcement contains no figures showing real-world operation or performance. Additional information is needed to assess its potential application to warehouse work.

The new element confirmed by this announcement is that RLWRLD presented its ‘RLDX-1’ robot foundation model in a live demonstration tied to a specific task: shipping work in a logistics warehouse. However, the supplied announcement describes a demo that “simulated” the task; it does not indicate that the model was tested in an operating warehouse or that it demonstrated how many packages it could process in a given period or its task success rate. It is therefore best viewed as a presentation linking the model to an envisaged task, rather than as evidence of its practical performance. Related articles in this newspaper include coverage of KDDI’s planned deployment of Rubin at its Osaka Sakai Data Center and expansion of inference infrastructure for physical AI (KDDI to deploy Rubin at Osaka Sakai Data Center, expand inference infrastructure for physical AI), as well as an article on robot deployment targets for a new company formed by KDDI and ELYZA (KDDI and ELYZA launch physical AI venture, aim to deploy more than 1,000 robots by fiscal 2030). Those reports concern computing infrastructure and robot deployment, while this announcement is about a demo of a specific task using an RLWRLD model; it does not establish that the efforts share a development structure or system. The supplied information also provides no basis for reading the multiple announcements involving KDDI as a single integrated plan. Structurally, a demo simulating logistics work appears to place an explanation of a foundation model in the context of a practical task. But without information about the robot itself, its gripper, perception sensors or control method, it is not possible to determine which parts of the task the model handles. Nor is it confirmed that this demo uses the same hardware or technology as the field trial covered in this newspaper’s article on a planned physical AI robot trial in Marunouchi by RLWRLD, Mitsubishi Estate and TMIP (RLWRLD, Mitsubishi Estate and TMIP plan physical AI robot trial in Marunouchi); the outcomes of the trials should not be conflated. Information to look for includes the specific steps demonstrated, the items and equipment involved, success rates across repeated runs, processing time, human intervention in exceptional cases, and whether testing took place in an actual warehouse. Such details would help clarify the distinction between presenting a foundation model and the robots, equipment and work design required for operational use.

According to the announcement’s headline and opening paragraph, RLWRLD presented a live demonstration using its ‘RLDX-1’ robot foundation model to simulate shipping work in a logistics warehouse. The supplied portion of the announcement does not specify the robot model, work procedure, demonstration date, performance metrics or whether the system has been used in actual operations.