SAP swaps costly AI models for cheaper alternatives, targets over €2 bln in gains
reuters.com · 28 October 2026
SAP replaced expensive AI models with cheaper alternatives in products including its internal consulting assistant. SAP targets more than €2 billion in AI-driven efficiency gains. SAP charges customers for completed tasks rather than for underlying computing resources. The company stated that the goal is healthy margins as scale grows, not an absolute reduction in AI expenditure. SAP uses more than 70 AI models from providers including OpenAI, Anthropic and Google, and increasingly chooses the cheapest model capable of performing a task to the required standard. Government agencies and companies in regulated industries increasingly ask specific requirements and exclude models based on origin. European customers increasingly demand local solutions.
This is the observable fact the cited source supports, stated without interpretation.
Testing CSD (capability scarcity) weakening: evidence shows SAP substituting more expensive proprietary models with cheaper alternatives, indicating reduced scarcity of capable compute. Testing LAW_2 (value moving from access to outcomes): SAP already prices by task outcome, not resource access. Testing LAW_1 (value moving toward orchestration): cost savings achieved by selecting lower-cost models suggests competing on model capability selection and orchestration rather than model production. Testing REVENUE_DURABILITY: substitution of inputs while maintaining margin goals tests whether task-based revenue models absorb input cost pressure.
SAP's transition from expensive to cheaper models while charging customers per task rather than per compute resource suggests the company is capturing cost savings at the service boundary while preserving gross task revenues. This is consistent with pressure on access-based pricing (capability supply now diversified across providers, cost of capability falling) and value moving toward orchestration and service design. The firm's stated goal of healthy margins at scale, not cost reduction alone, indicates awareness that volume scaling of commodity-model-backed services faces margin pressure unless value reaccumulates in orchestration and customer outcome certainty. The emerging customer exclusions based on origin and demands for local solutions establish a new scarcity axis independent of model capability itself.
SAP's transition from expensive to cheaper models while charging customers per task rather than per compute resource suggests the company is capturing cost savings at the service boundary while preserving gross task revenues. This is consistent with pressure on access-based pricing (capability supply now diversified across providers, cost of capability falling) and value moving toward orchestration and service design. The firm's stated goal of healthy margins at scale, not cost reduction alone, indicates awareness that volume scaling of commodity-model-backed services faces margin pressure unless value reaccumulates in orchestration and customer outcome certainty. The emerging customer exclusions based on origin and demands for local solutions establish a new scarcity axis independent of model capability itself.
This is the framework's reading of the fact above. The source does not endorse it and is not cited as doing so.
The observation captures an active substitution of model input costs in October 2026, after the April 2026 paper publication. SAP's explicit goal statement about margins at scale is a direct signal of value structure pressure. Whether this translates to a 1 to 2 year durability test on revenue models remains longitudinal; the current evidence shows the substitution is happening and revenue model redesign (task pricing versus resource pricing) is already in place. The emerging geographic and regulatory exclusions add a new scarcity dimension that may widen over the next 2 to 3 years.
Current evidence is recorded immediately. The paper's longer empirical horizon determines when persistence becomes a stronger framework-level test.
SAP may simply be optimising its cost structure during a period of model commodification without facing any fundamental margin pressure. If competitors face identical cost reductions, competitive dynamics may preserve margins industry-wide. SAP's efficiency claim (€2bn) is the company's own projection, not independently measured outcome.
Recorded for every observation assessed at strength 4 or 5. A record that cannot state the strongest competing reading of its own evidence is not published.
Where several outlets report the same development, they are recorded as additional sources on this one observation and are not counted as separate evidence.