Autonomous Enterprise: Model Forecast for 2026–2029

A Tetlock superforecasting framework, calibrated to Autonomous Enterprise strategy
Vol. I · No. 5
July 2026
3 outcomes · 5 parameters
Three scenarios, anchored to four prior SAP platform shifts (PC, internet, cloud, analytics). Adjust the parameters as evidence arrives — probabilities update live via softmax over weighted log-odds. The red tick marks the base rate; the bar is the current posterior. A tool for structured belief-tracking — not a prediction or recommendation.
Entropy
—
Modal
Concentration
—
Drift From Base
—
Enter a business email above to enable updates.
Current
Vol. I · No. 5
July 2026
Lead27%
Defend50%
Consolidate23%
The Defend scenario reached majority probability for the first time. Two structural facts drove the move: only 3% of SAP customers run Business AI in production (DSAG Investment Survey 2026), and an estimated 40–45% of the ECC base is projected to remain on-premise at the 2027 deadline — structurally excluded from Joule entirely. The free window on Joule Studio and agent runtime closes end-2026; 2027 pricing for agent throughput and data access remains undisclosed. Agentforce has expanded into operational workflows — historically SAP territory — and analysts rate Joule use cases as less impressive in head-to-head comparisons. Partial offset: named marquee customers (RWE, Ericsson, KPMG) and the SAP–n8n investment signal real ecosystem commitment.
Anthropic partnership stability 0 (neutral) → 0 No change. Partnership intact; concentration risk flagged by Forrester but no adverse signal materialised.
Agent Hub cross-vendor parity 0 → 0 Agent Hub GA still Q3 2026. 680+ partner submissions positive, but parity in production unverified.
Operational data substrate 0 → −1 BDC GA slipped to H2 2026. DSAG Hamburg: "Business Data Complexity." Integration docs inconsistent. Reltio close and HANA Cloud native partially offset.
Installed-base AI activation pace 0 → −2 3% production adoption (DSAG). 77% of AI-active SAP enterprises using non-SAP tools. 40–45% ECC customers expected on-premise at deadline. Joule structurally locked to RISE/GROW.
Path 1 commitment among buyers 0 → −1 Agentforce expanding into back-office. Joule demo parity below Agentforce per analyst reports. Path 3 (AI-native verticals) gaining credibility.
Vol. I · No. 4
May 2026
Lead32%
Defend44%
Consolidate24%
Post-Sapphire 2026 update. SAP unveiled the Autonomous Enterprise framework with 50+ Joule Assistants, 200+ specialized agents, and the SAP Business AI Platform consolidating BTP, BDC, and AI Foundation. The SAP–Anthropic partnership was announced as the model layer. Forrester flagged concentration risk; OSS Ventures noted a 4% market-cap drop on the day. Defend retained modal status. Lead rose on the strength of the Agent Hub announcement and the Anthropic stack — both dependent on cross-vendor parity delivering in production.
Anthropic partnership stability +1 introduced SAP–Anthropic deal announced at Sapphire. Forrester flagged as single-vendor concentration risk.
Agent Hub cross-vendor parity 0 introduced Announced at Sapphire; Q3 2026 GA. 680+ partner submissions. Evidence pending.
Operational data substrate 0 introduced BDC positioned as the data foundation. OSS Ventures noted ~40% of agentic-decision data sits outside structured SAP systems.
Installed-base AI activation pace 0 introduced Autonomous Close Assistant the lead revenue signal. No production adoption data at time of issue.
Path 1 commitment among buyers 0 introduced a16z Path 1/2/3 framing introduced as the key structural lens for buyer behaviour.
Vol. I · No. 3
March 2026
Lead26%
Defend48%
Consolidate26%
Pre-Sapphire calibration. DSAG Technology Days 2026 in Hamburg surfaced friction: BDC documentation gaps, customer spending described as increasingly selective, and AI investment going to non-SAP tools. DSAG Investment Report 2026 released (February). Defend reinforced; Consolidate held. Lead edged down on the evidence that customer AI spend was routing around SAP's native stack.
BDC adoption velocity +1 → 0 DSAG Hamburg: "Business Data Complexity." Documentation inconsistencies reported. Moved from mildly positive to neutral.
Joule competitive performance 0 → −1 DSAG Investment Report 2026: only 3% of customers running SAP Business AI in production. 77% using non-SAP tools.
Vol. I · No. 2
January 2026
Lead28%
Defend46%
Consolidate26%
Q4 2025 signals incorporated. BDC Connect for Databricks reached GA; Snowflake and BigQuery scheduled for H1 2026. SAPinsider benchmark: 30% of respondents fully live on S/4HANA Cloud Private (up from 19% the prior year). Migration momentum mildly positive. Joule rollout to 35 solutions but enterprise production adoption data not yet available. Base rates confirmed from the four-platform-shift reference class.
BDC adoption velocity 0 → +1 BDC Connect for Databricks GA. Snowflake/BigQuery H1 2026 on track. Open ecosystem signal positive.
RISE migration rate 0 → −1 ~39% of ECC customers licensed S/4HANA as of end-2024. SAPinsider: migration linear, not accelerating. ECC deadline pressure building but not yet panic-rate adoption.
Vol. I · No. 1
October 2025
Lead19%
Defend42%
Consolidate39%
Inaugural issue. Base rates anchored to four prior SAP platform shifts: PC (uncontested leadership), internet (core held; HCM/CRM repositioned to focused vendors), cloud (multi-year transition via RISE), analytics (share moved to cloud-data-lake architectures; Dremio acquired as hedge). Base rate for uncontested leadership: roughly 1-in-4. All five parameters initialized at neutral pending evidence accumulation. Defend set as modal — the historically dominant pattern when SAP's architecture diverges from the market's adjacent-layer trajectory.
BDC adoption velocity 0 (initialized) BDC announced Feb 2025. No production adoption evidence at launch.
RISE migration rate vs 2027 0 (initialized) ECC deadline confirmed. Migration progress steady but slow.
Joule competitive performance 0 (initialized) Joule launched 2023. Competitive position vs Agentforce/Copilot unresolved at time of issue.
Marquee customer concentration 0 (initialized) No NDA-lifted production reference customers at time of launch.
EU Digital Markets Act 0 (initialized) DMA applicable from May 2023. SAP not a designated gatekeeper. Cloud designation investigation pending.
§ § §

Outcomes

Lead — SAP extends from system-of-record into the agent operating layer, in production at scale. Defend — SAP holds its ERP and financial-close core; the agent layer goes to other platforms (the historically modal outcome). Consolidate — extended adoption timelines or share recalibration in the agent layer, echoing the 2010–2024 analytics shift.

Methodology

Base rates anchor to four prior SAP platform shifts: PC (uncontested leadership, after a slow start), internet (core held; HCM and CRM repositioned to focused vendors), cloud (multi-year transition, re-activated via RISE), analytics (share moved to cloud-data-lake architectures; Dremio acquired as a hedge). Pattern: uncontested leadership is roughly 1-in-4; adjacencies reposition when SAP's architecture diverges from the market's.

Each parameter carries a weight vector across the three scenarios. The Likert value {−2…+2} multiplies the weight; contributions add to base log-odds; softmax normalizes to 100%.

Use

A scoring rubric, not a forecast. Move a parameter only on specific, recorded evidence — an analyst note, a reference call, a field observation. Log the evidence with each change.

References — Vol. I · No. 5 (July 2026)

DSAG Investment Survey 2026 (February 2026) — 3% production adoption of SAP Business AI; 77% of AI-active SAP enterprises using non-SAP tools.

DSAG Technology Days 2026, Hamburg — "Business Data Complexity" reinterpretation; BDC documentation gaps reported by Michael Bloch, Head of Licenses, Contracts and Support.

Innobu, "SAP Joule 2026: Agentic Enterprise AI — Promise vs. Reality" (April 2026). Source.

IgniteSAP, "SAP Joule Adoption and What It Means for Your Career" (June 2026). Source. — Forrester: of 224 agents and 51 assistants announced at Sapphire 2026, most in EAC or preview rather than full GA.

Constellation Research, Larry Dignan, "DSAG: SAP customer spending selective, AI drags" (February 2026). Source.

SAVIC Technologies, "SAP ECC Migration 2026: Last Chance Before 2027 Deadline" (May 2026) — ~39% of ECC customers licensed S/4HANA; 40–45% projected to remain on ECC at deadline. Source.

SAP News Center, "Accelerate the Autonomous Enterprise with SAP Business Data Cloud" (May 2026) — BDC GA planned H2 2026; Reltio integration completed; BDC Connect for Amazon Athena announced. Source.

SAVIC Technologies, "SAP Invests in n8n at $5.2B" (May 2026) — SAP $60M investment; n8n integration in Joule Studio planned Q3 2026. Source.

Forbes / Victor Dey, "The End of the ERP Era?" (May 2026) — KPMG: 270,000 Joule users company-wide, 3,000 consultants actively using 20 agents. RWE offshore wind deployment cited. Source.

Nanonets, "SAP Sapphire 2026: The Complete Breakdown" (May 2026) — Agent layer described as "real but immature"; Agentforce parallel noted. Source.

References — Vol. I · No. 4 (May 2026)

Forrester, F. Medhora, "SAP Sapphire 2026: The Autonomous Enterprise Is Credible, But It Comes With Concentration Risk" (12 May 2026). Source.

OSS Ventures, R. Devillieres, "SAP Sapphire: cool slides, 4% drop in market cap" (15 May 2026). Source.

a16z, A. Strange — enterprise AI architectural paths (Path 1 / 2 / 3), May 2026. · Goldman Sachs commentary on the SAP–Anthropic announcement (12–13 May 2026). · SAP News Center, Sapphire 2026 Keynote and Innovation News Guide. Source.

Framing

Built for structured belief-updating, not to advocate for or against any platform. The reference class is observed pattern, not projection. Assign your own weights as evidence accumulates in your context.

Opening your email client…