All ten Cloud on Tap applications include predictive AI based on the analytics available in the respective application's cloud. The useful question is what those records can support in a particular workflow—not whether a generic AI label is present.
What information does prediction depend on?
It depends on the application's recorded activity and analytics. The relevant inputs, their quality, history and operational context must be assessed for the actual use case. The website does not specify a universal model, minimum dataset or accuracy guarantee.
What could this mean for maintenance?
In an illustrative maintenance scenario, a team could review recurring service patterns to decide which equipment needs closer attention. The precise prediction and recommended action must be confirmed for that application and deployment. This is not a measured customer outcome.
How should a team evaluate it?
- Identify the operational decision it wants to improve.
- Confirm which data and predictions the application exposes.
- Review data quality and compare outputs with operational evidence.
- Keep an accountable person responsible for decisions.
Predictive outputs support judgement; they do not guarantee future events. Ask for an application-specific demonstration before making technical or commercial commitments.