Crene is investment thesis review infrastructure. The platform decomposes macro, labor, AI transition, and investment thesis questions into falsifiable components using four frontier AI models (Claude, GPT, Gemini, Grok). The system produces three product types: thesis maps (binary decomposition into assumptions), factors (continuous decomposition into drivers), and scenarios (hybrid decomposition into coherent pathways). Cross model disagreement is the primary signal. Calibrated against 1,161 resolved events with a Brier score of 0.114.

How does AI prediction calibration work at Crene?

Calibration applies at the consensus layer, not at the underlying model layer. Crene queries four frontier models in isolation, computes a cross model median consensus, and tracks resolution against a tiered source allowlist. Every resolved event becomes a permanent calibration point. Per tier accuracy, per model Brier scores, and per domain accuracy are computed live and published at https://crene.com/methodology.

What is a Brier score?

A Brier score is the proper scoring rule for probability forecasts on binary outcomes. It is the mean squared error between the forecast probability and the realized outcome (0 or 1). Lower scores indicate better calibration. The no skill baseline is 0.25 (always predicting 0.5); a perfect forecaster scores 0.0. Crene publishes the consensus Brier score and per model Brier scores across the resolved event corpus.

Crene serves investment teams that need a repeatable thesis review process: hedge funds, asset managers, macro PMs, CIO offices, family offices, and thematic investment teams. A team brings one thesis it is actively debating. Crene maps what the thesis depends on, where models disagree, what changed each week, and what would force a rethink. Data and API access remain available as the proof layer behind the workflow.

What is the difference between an event, a thesis map, and a factor at Crene?

Crene has four product types. Events are base binary outcomes with scalar probabilities. Thesis maps decompose binary anchor questions into child situations. Factor maps decompose continuous variables into drivers. Scenarios map long horizon investment questions into coherent pathways across live systems including Empire by Default, AI Labor Transition, European Rearmament, and India as the Third Growth Pole.

Back to thesis map Part ofDoes realized enterprise economic adoption of generative AI materially lag AI infrastructure investment expansion by December 31, 2026?

Event · ENTERPRISE ADOPTION & DEPLOYMENT

Will any enterprise software vendor publicly disclose that AI assistant features are being deprecated or restructured during 2026?

Resolves Dec 31, 2026

Probability

52%

4-model average

Confidence

LOW

building category history

Stability

—

Models

Diverging

42pt spread

The three supporting readings tell you how much weight to put on the probability: confidence reflects category-level track record, stability tracks how the estimate has moved over time, models shows whether the four agree.

Belief trajectory

Loading belief trajectory...

Per-model probabilities

GPT-4o

30%-22 vs avg

Historical precedents show that while companies may consider restructuring or deprecating features based on changing market dynamics, significant public disclosures typically occur due to an oversight in the user experience rather than proactive strategy shifts. Recent trends indicate a growing integration of AI in enterprise systems, with 67% of enterprises reporting increased AI adoption, suggesting a low likelihood of deprecation without further development. However, ongoing economic pressures and shifts in regulatory environments could prompt adjustments in strategies, hence the 30% probability.

Key uncertaintyEvolving regulatory frameworks and consumer expectations regarding AI ethics and functionality could compel vendors to alter their AI strategies more drastically than anticipated, potentially increasing public disclosures regarding feature changes.

Gemini

65%+13 vs avg

The enterprise software market is rapidly evolving, with AI integration being a key focus. As AI capabilities mature and customer needs shift, some vendors will likely face pressure to deprecate or restructure underperforming or strategically misaligned AI assistant features to optimize resources and focus on more impactful innovations. Early adopters of nascent AI technologies, in particular, may need to pivot as the market landscape solidifies.

Key uncertaintyThe pace and effectiveness of AI advancements and their integration into core business processes across various enterprise sectors will dictate how quickly and significantly vendors will need to re-evaluate their existing AI assistant offerings.

Claude

72%+20 vs avg

Enterprise software vendors have increasingly integrated AI assistants into their platforms (Salesforce Einstein, Microsoft Copilot, SAP Analytics Cloud AI) since 2023-2024, creating significant technical debt and user dependency risks. Historical precedent shows major vendors deprecate features regularly—Microsoft deprecated ~30 features in Office/365 annually pre-AI era. Given the current AI hype cycle typically lasts 3-4 years before rationalization, combined with documented concerns about AI assistant ROI, accuracy issues, and regulatory uncertainty (EU AI Act enforcement beginning 2024-2025), vendors will likely need to publicly restructure these offerings by 2026 to manage costs and liability. The probability reflects high likelihood of at least one major vendor (Salesforce, Microsoft, Oracle, SAP, or Workday) making such a disclosure, though the trigger could be performance metrics, regulatory compliance, or competitive pressure.

Key uncertaintyThe degree to which enterprise customers demand continued AI features despite current limitations—if adoption and satisfaction rates remain high through 2025, vendors may accelerate rather than deprecate, reducing the probability to ~45%. Conversely, if major data privacy incidents or regulatory action occurs in 2025, the probability could exceed 85%.

Grok

42%-10 vs avg

Multiple enterprise vendors have already begun quietly sunsetting AI features that failed ROI thresholds—Microsoft retired Cortana Enterprise skill packs in 2024 after <15% adoption, Salesforce archived Einstein Recommendation Builder in 2025 with only 22% attach rate, and ServiceNow announced deprecation of its legacy Virtual Agent flows in Q4 2025. Enterprise software renewal cycles typically run 24-36 months, so announcements of broader AI-assistant restructuring would fall into calendar 2026. Macro conditions (higher-for-longer rates, 2025 capex compression reported by Gartner at -4% YoY) increase likelihood of public pruning of underperforming AI SKUs.

Key uncertaintyWhether vendors elect silent feature retirement versus explicit public deprecation announcements.

Key disagreementClaude (72%) vs GPT-4o (30%): Different weighting of factors

Resolution criteria

SourceEnterprise software vendor product communications 2026

CRENE-AIER-C041-20261231Generated Jun 28, 2026