SAP has launched its Autonomous Enterprise platform with AI agents performing business tasks autonomously and shifted to a novel pricing model that bills customers per 'agent action.' Industry analysts caution this approach could lead to unexpectedly high expenses due to ambiguous usage metrics and evolving contract terms.
- SAP bills AI agent use by completed actions, not user licenses.
- Ambiguous 'action' metrics may cause spiraling and unpredictable costs.
- Gartner advises customers to review contracts and conversion rates closely.
What happened
SAP introduced its Autonomous Enterprise platform incorporating AI agents designed to autonomously execute business tasks. This is part of SAP’s broader effort to accelerate AI adoption within its cloud services and migrate customers away from legacy software.
In parallel with the platform launch, SAP unveiled a new commercial model that charges customers based on the number of completed 'actions' performed by AI agents rather than traditional licensing metrics tied to user counts. The vendor is developing Autonomous Domain Blueprints to guide customers in estimating costs based on deployment scale, though key billing definitions remain unclear.
Why it matters
The shift to an action-based billing model creates uncertainty for enterprises about how quickly costs might accumulate since the exact definition of a billable 'action' has not been detailed. This ambiguity raises the risk that charges could escalate unexpectedly, especially if unit prices increase or if AI agents consume additional digital licenses.
Gartner’s analysis highlights that contract flexibility allowing SAP to adjust conversion rates of AI Units at renewal further complicates cost predictability. For large SAP customers like Walmart and Volkswagen relying on these AI capabilities across critical operations, unforeseen cost surges may impact budgets and vendor relationships.
What to watch next
Enterprises considering SAP’s AI platform should proactively review existing contracts for price protection clauses tied to SAP Cloud applications such as S/4HANA. They should request and analyze the latest AI Services List and conversion factors from SAP’s Trust Center to establish a baseline for AI Unit consumption costs.
SAP’s forthcoming Autonomous Domain Blueprints and disclosed runtime metrics for their Joule Studio agent builder platform will be critical for customers to better forecast ongoing expenses. Industry watchers will track updates from Gartner and SAP for clarifications on pricing definitions and policies as adoption grows.