Previously tedious, slow, and bureaucratic processes involving the allocation and administration of computational resources can be reimagined now that artificial-intelligence systems can use natural-language and perform reasoning tasks.

Authenticated human users and artificial-intelligence systems requesting permissions to access computational resources could complete input forms and/or engage in natural-language dialogues to do so. As considered, these input forms and dialogues would involve descriptions of their tasks, goals, plans, and strategies. Presented here is that intelligent resource allocation and administration systems could analyze such request data to better allocate and schedule concurrent access to computational resources.

For a first example, let us consider a supercomputer. In order to request permissions to access and use it, both authenticated human users and artificial-intelligence agents could complete input forms or engage in natural-language dialogues to describe their tasks, goals, plans, and strategies. These request data could be subsequently used by intelligent resource allocation and administration systems to reason about, allocate, and schedule concurrent access to computational resources.

For a second example, let us consider a large-scale artificial-intelligence system. As considered, access to such a resource by authenticated human users and smaller-scale artificial-intelligence agents would require permissions. By including tasks, goals, plans, and strategies in permission requests, quality of service could be greatly enhanced and, also importantly, secondary components could ensure that all primary interactions remained "on task", "on track", and "in scope".

That is, smaller, secondary artificial-intelligence agents and any in-the-loop human operators could analyze those input data provided in permission requests, allocate and schedule access based upon these analyses, and monitor primary interactions to ensure that they unfolded as described and subsequently permitted. Completed input forms or dialogues could, effectively, be used to create and modify structured documents detailing parameters of expected or allowed interactions with large-scale artificial-intelligence systems. As considered, these input forms or natural-language dialogues could be completed once per authentication session, period of time, or granular task.

In conclusion, intelligent resource allocation and administration systems could use input data provided in permission requests, these input data including tasks, goals, plans, and strategies, to enable more efficient and faster-paced stewarding of computational resources by human personnel.