KIDZ AI signs a non-binding data center MOU to assess a GPU-focused AI inference deployment tied to a $44.6 million Canopy Wave agreement

KIDZ AI has entered into a non-binding memorandum of understanding with Limestone Networks to evaluate deploying AI inference infrastructure at a Dallas data center, according to a market update posted Tuesday. The agreement is positioned as an operational step connected to a larger Canopy Wave GPU compute arrangement valued at $44.6 million.

In practical terms, an AI inference deployment refers to the use of trained AI models to generate outputs in real time or near real time, such as answering queries or processing requests, rather than training new models from scratch. For most inference systems, compute hardware such as GPUs is central because it can accelerate the heavy math involved in running modern machine-learning models.

The memorandum of understanding is described as non-binding, meaning it sets out an evaluation plan rather than final contractual terms for construction, procurement, or ongoing compute services. Still, the company framed the Dallas data center evaluation as part of advancing toward execution under its Canopy Wave GPU compute agreement.

The market update did not provide additional granular details about what infrastructure would be installed at the site, the timelines for completing the evaluation, or the specific performance or commercial criteria that would determine whether the parties move from study to a binding agreement. It also did not clarify how the $44.6 million figure would be structured in operational terms, such as pricing by compute usage, a fixed service fee, or milestone-based components.

Limestone Networks, in this arrangement, is the partner responsible for hosting or enabling the data center environment for the proposed inference infrastructure. The company’s next steps, as described, revolve around determining whether a deployment at the Dallas facility is feasible and operationally suitable.

The development lands in a wider market context in which AI hardware capacity and data center access have become key bottlenecks for companies attempting to commercialize AI services. Inference-focused projects often turn on the ability to secure reliable compute in the locations where customers and operational teams can support rollout, testing, and ongoing service delivery.

A key caveat is that the update does not disclose whether any binding contract will follow the MOU, how the parties would measure readiness for deployment, or what specific GPU models or vendor stack would be used. It also does not state whether the Canopy Wave arrangement is already active, only in planning, or contingent on completing the Dallas evaluation.