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.

Bring your own thesis

Get this same map built on a thesis your team is debating.

Assumptions, model disagreement, and rethink triggers, updated weekly before your PM, risk, or IC discussion. Currently accepting one macro thesis and one AI-economy thesis for July.

Map one thesis

Back to thesis map Part ofWill AI generated content exceed human generated content on the US internet before 2030?

Event · LEGAL, COPYRIGHT & REGULATION

Will any US state ban AI generated content in public school educational materials by 2029?

Resolves Dec 31, 2029

Probability

30%

4-model average

Confidence

LOW

building category history

Stability

—

Models

Mixed

17pt spread

Current read

Consensus sits at 30% across the four models and is — (loading). Models are mixed at a 17pt spread, so the average conceals live disagreement worth inspecting below. Category confidence is LOW (building category history).

Confidence reflects category-level track record. Stability tracks estimate movement. Models shows whether the four agree.

Belief trajectory

Loading belief trajectory...

Per-model probabilities

GPT-4o

30%

Several states have recently proposed legislation addressing educational content and data privacy, reflecting a growing concern over AI’s role in society. For example, the recent push for curriculum transparency and regulation, as seen in states like Florida and Texas, signals a willingness to scrutinize the materials used in education, which could extend to AI-generated content. Furthermore, the debate around the ethical implications of AI, particularly in educational settings, is likely to increase as awareness of AI's potential biases and inaccuracies grows, influencing state policies by 2029.

Key uncertaintyThe reaction from education stakeholders, including teachers, parents, and technology firms, could sway legislation either in favor of banning AI-generated content or promoting its responsible use, affecting the likelihood of such bans materializing.

Gemini

40%+10 vs avg

Growing concerns about the accuracy, bias, and copyright of AI-generated content in educational settings are leading to legislative scrutiny. As of early 2024, several states are discussing or have introduced legislation related to AI governance, though none have specifically targeted outright bans in educational materials for public schools. The rapid evolution of AI and the increasing integration of such tools into curriculum development present a plausible, though not yet certain, path to specific bans within this timeframe.

Key uncertaintyThe development of robust AI detection and content verification tools that can be effectively and affordably implemented by school districts to identify and flag AI-generated materials.

Claude

28%-2 vs avg

While AI regulation in education is accelerating (several states introduced AI literacy bills in 2023-2024, and the Biden administration issued AI executive orders), an outright ban on AI-generated content in educational materials by 2029 faces significant headwinds. Historical precedent shows US states rarely implement blanket content bans (textbook adoption policies are content-neutral); instead, they prefer disclosure/transparency requirements or curriculum guidelines. Current momentum favors AI integration in schools (major ed-tech platforms are adding AI features with state adoption), and a 5-year timeline is relatively short for legislative consensus on such a restrictive measure. However, rising concerns about academic integrity, misinformation, and child safety could accelerate regulatory action in 2-3 states.

Key uncertaintyWhether high-profile AI incidents in education (e.g., widespread plagiarism scandals or documented harms to students) occur in the next 2 years, which could dramatically shift political will toward bans rather than lighter-touch regulation.

Grok

23%-7 vs avg

Four states (California, New York, Texas, Florida) already introduced 2024 bills restricting AI in K-12 materials; none passed outright bans but three enacted disclosure rules (CA AB 2839, NY S07686, TX HB 1848). Current K-12 textbook adoption cycles run 5–7 years, so any 2025–2027 statute would affect materials purchased for 2028–2029; no state education agency has yet published AI-content standards. Base rate for technology-specific curriculum bans is low: only 2 of 50 states banned calculator use in math classes post-1975, both reversed within 4 years.

Key uncertaintyWhether federal Title IV or IDEA funding rules will explicitly classify AI-generated text/images as allowable “instructional materials,” which would pre-empt most state-level prohibitions.

CRENE-AICD-LR05Generated Jul 2, 2026