SAP Puts Its Tabletop AI Model in Orbit While Buyback Ledger Keeps Filling

SAP has found an unusual proving ground for its artificial intelligence ambitions: low Earth orbit. The German software group plans to deploy SAP-RPT, its tabular AI model, aboard the International Space Station in early 2027, running on Hewlett Packard Enterprise's Spaceborne Computer-2. The stated goals are real-time anomaly detection and forecasting for telemetry data — a technical showcase far removed from the enterprise sales cycle.

Closer to the balance sheet, the company's financial tooling is moving at a more conventional pace. SAP Pay is now generally available in the US and UK, according to the company. SAP Enterprise Financial Consolidation is slated to enter its "Early Adopter Care" phase in the first quarter of 2027 — meaning the payments product is live in those two markets while the consolidation solution remains a promised step.

Buybacks and a Backlog Under Scrutiny

An interim disclosure published Tuesday put SAP's repurchases for October 26 through October 30 at 50,000 of its own shares, all acquired via Xetra. Cumulative purchases under the current tranche have reached 9,136,956 shares as of the end of that window.

The operational headline came roughly two weeks ago with quarterly results: the current cloud backlog rose 28% to EUR 24.1 billion in the third quarter of 2026. SAP also guided cloud revenue for 2026, at constant currencies, toward the upper end of its EUR 25.8 billion to EUR 26.2 billion range. Preliminary figures for the fourth quarter and the full 2026 fiscal year are due on January 27, 2027.

The Real Test Is Attach Rate, Not Announcements

For investors, the next chapter of the AI story hinges on commercial absorption rather than product launches. The key gauge is how deeply Business AI is attached to RISE with SAP and S/4HANA cloud contracts. If that attach rate climbs while margins hold, it would signal successful monetization. SAP Document and Reporting Compliance is also gaining traction across several European jurisdictions, according to media reports — most meaningful for shareholders if broader adoption translates into incremental revenue.

Partner NTT DATA reports measurable gains among European enterprises in process automation and decision support. Whether that translates into additional contract wins and stable margins is the question that matters. A week or so ago, SAP and NTT announced a planned collaboration on SAP Sovereign Cloud and SAP Business AI, initially linking on-site sovereign cloud with infrastructure and operations managed by NTT, with further AI use cases in regulated industries under review.

Copilot Outruns Joule in the Installed Base

The competitive picture is less comfortable. A study published Wednesday by Precisely and ASUG found Microsoft Copilot in use within SAP processes at 60% of surveyed companies, against 26% for SAP Joule. That gap points to robust demand for AI assistance around SAP workflows — an economic opportunity for SAP only if it can anchor its own offerings more firmly in customer contracts. Copilot's wider footprint also raises the pressure to demonstrate Joule's distinct value in day-to-day work.

The bull case links successful customer deployments with deeper contract penetration: if automation and decision support deliver demonstrable value, more customers could adopt additional SAP AI products, provided revenue outpaces the associated costs. A growing cloud base helps — total revenue surpassed EUR 10 billion for the first time, and SAP raised its 2026 outlook. One source put third-quarter cloud backlog growth at 26%; the more recent quarterly disclosure cites 28%, the figure to use.

Margin Risk and a Cautious Company Tone

The bear case starts with value capture. If companies mostly run AI in SAP processes through third-party providers, SAP collects a smaller slice of the incremental willingness to pay — precisely the risk Copilot's lead illustrates. Cloud infrastructure and digital sovereignty costs could squeeze margins if SAP cannot pass them through in pricing.

Management's own commentary warrants attention: SAP describes the business climate as rather subdued and sees the effects of US interest rates becoming more pronounced. That weakens the optimistic scenario if customers defer additional spending while delivery costs rise. Weak attach rates combined with margin pressure would be a serious warning signal — a large catalog of available applications only convinces economically when customers use them and SAP grows profitably as a result.

What to Watch Before Year-End

The next concrete checkpoint is the expansion of the AI portfolio targeted for the end of 2026 — internal delivery goals that matter to investors mainly if broader commercial adoption follows. The fourth quarter carries extra weight, with SAP calling it decisive for the full-year result. The most reliable yardstick remains the combination of usage, contract business and profitability: growth quality should count for as much as its pace.