2026 product definition
Plug any model in as the brain. KELLY forces: ask → missing facts → forecast/answer → proof receipt → check-in → settlement → calibration → only then a governed memory proposal.
Sector labs are lenses. JobProof is paused. Brains are engines — not the product. Ops ceilings (Neon opt-in, soft-cap) are not densify thrash — after Railway ship: npm run accountability:post-deploy-hold.
Truth console · public scoreboard
Live public Brier volume mapped to the loop. Process is not world truth. Memory remains gated.
Settled
0
Weather / Sports
0 / 0warming
Markets
0
Scale remaining
1000
Causalea public home · Answered directly by KELLY · public writes blocked
KELLY LLM utilization (engine policy)
Still sensible: LLM as communication adapter only
Truth debt · process only
Loading process debt…
Truth debt tracks unfinished proof / product legs — not settled world outcomes.
Claim boundary · process only
Loading claim boundary…
Memory writes: blocked · Public writes: blocked
KELLY is the missing layer: ask for missing facts, issue a proof receipt, check reality later, score calibration, and only then propose earned memory. Causalea is the public home and professional lenses around that loop.
www.causalea.com
Public home of KELLY
Causalea is the doorway. KELLY is the engine that answers every serious lens.
KELLY
Decision engine
Bounded answers · dual-lane forecasts · settlement · calibration · governed memory
Causalea
Interactive lenses
Ask · desks · Glass Lab proof · professional sector workplaces
Ask
decide
Forecast
score
Proof
audit
Memory
lineage
What you are looking at
First-time visitors often ask whether Causalea is “another AI chatbot.” It is not. Here is the split.
KELLY
KELLY is the organism that reasons under uncertainty: dual-process discipline, causal claims, forecasts that can be scored, settlement against reality, calibration, and governed memory with lineage.
Causalea
Causalea (this site) is the professional surface people use: landing page, sector desks, Ask, Superforecasting, and Glass Lab. Every serious path is answered by KELLY — Causalea is not a second brain.
What you can do
You do not need to learn the whole organism on day one. Pick a job, open a lens, and ask KELLY.
Ask a real decision question. Get a bounded answer: what is known, what is uncertain, what to do next, and where proof lives.
Ask a question
Live odds → implied probability → KELLY disagreement → settlement score. A paid sports analytics desk — not a betting slip.
Open odds desk
Place-aware hazard windows with feed freshness and observed closure. Situational awareness professionals pay for.
Open hazard desk
Public Brier, reliability, and first-win volume. The product customers audit before they renew.
View scoreboard
How it works
KELLY is built for decisions under uncertainty — where “a confident paragraph” is not enough.
01
Ask-KELLY is the front door. KELLY asks for missing context before a serious read when stakes need proof.
02
Not polish alone: assumptions, confidence, flip conditions, evidence readiness, and what would prove it wrong.
03
Check-ins and public scoreboard Brier close the loop. Calibration tracks whether 70% claims were right ~70% of the time.
04
Chat never becomes truth. Only after proof + outcome + governance may memory be proposed. Glass Lab is the truth console.
Why KELLY is different
Most AI products optimize for a polished reply. KELLY optimizes for decisions you can audit: sources, uncertainty, settlement, calibration, and memory only after reality and court agree.
Serious answers point to sources, receipts, replay, or an honest waiting state — not just fluent text.
Your probability and KELLY’s stay separate so calibration stays honest after settlement.
Bounded next steps under uncertainty — including when the smart move is to wait or probe.
You can ask and inspect. Only KELLY’s governed paths write memory and ground truth.
Who it is for
Need a clear next step, risks, and what would change the answer — without fake certainty.
Open path →
Need dual lanes, settlement, and calibration — not blended confidence theater.
Open path →
Need receipts, generation wall, truth debt, and a no-write public boundary.
Open path →
Markets, hazards, policy, science, and memory each get a workplace lens — same KELLY engine.
Open path →
Sector desks
Each desk is an interactive lens with its own instruments — still answered by one engine.
Decision advisor
Turn a plain-language question into a bounded decision: act, wait, or test — with uncertainty and proof one click away.
Superforecaster
Frame measurable events, pre-commit dual probabilities, settle against reality, and score calibration.
Market analyst
Read quote tape with freshness first, forecast cones with uncertainty, paper score — never trade execution.
Proof auditor
Public trust console: receipts, replay, organ health, truth debt, generation wall, and no-write boundary.
Sports probability analyst
Convert live book lines to fair implied probability, measure KELLY disagreement, score after game close — never place bets for the user.
Hazard intelligence officer
Place-aware weather and disaster windows with official source freshness, alerts, and observed closure — not emergency command authority.
Policy & records researcher
Start with official citations and effective dates, then macro signals — never partisan framing or legal advice.
Research evidence auditor
Map claim support, contradiction, recency, and falsifiers — evidence quality over volume.
Theory builder
Turn observations into falsifiable questions with source plans and closure targets.
Memory lineage archivist
Trace birth-to-root ancestry, generation walls, supersession, quarantine, and repair eligibility.
Paid-grade accountability · evidence-first product
Customers pay for desks that settle: sports probability, hazard intelligence, and scored forecasts. Open forecasts wait on reality. Honest empty beats fake Brier.
0
0
waiting
Proof lane 0/30 settled · open+settled scale target 1000
Live from KELLY · memory provenance
Root cells, gen-1 revision children, and ancestry edges — inspected through Causalea lenses, owned by KELLY.
3,247
13
13
Trust & transparency
When an answer matters, open Glass Lab: package truth, receipts, generation wall, truth debt, accountability bind, organ health, and the public write boundary. If something is incomplete, the interface should say so.
Receipts
Proof that a process ran
Settlement
Reality closed a claim
Calibration
Skill after outcomes
Memory lineage
Who begot which cell
Ready when you are
Ask KELLY something you actually care about — with stakes, horizon, and what would settle the answer. Then open the proof path if the decision matters.
www.causalea.com · Causalea lenses · KELLY engine · public writes blocked
