Compound Fork — AI pause × Compute scale
Each cell = a joint scenario "Both branches fire". Cell intensity = total |Δ| across the 200 highest-conviction tradeable predictions vs their unconditional posterior. Joint probabilities approximated via log-odds combination (assumes conditional independence given the prediction — coarse but bounded). Click a cell to drill into the dominant single-fork view.
Pick two fork families
| AI pause ↓ × Compute scale → | COMPUTE_1GW_2027 prior 60% | COMPUTE_10GW_2028 prior 40% | COMPUTE_100GW_2030 prior 20% | COMPUTE_STARGATE_FAILURE prior 15% |
|---|---|---|---|---|
AI_PAUSE_2026 prior 5% | 125 claims · Σ|Δ| 16.18 | 129 claims · Σ|Δ| 17.27 | 127 claims · Σ|Δ| 17.77 | 124 claims · Σ|Δ| 16.08 |
AI_PAUSE_2027 prior 10% | 126 claims · Σ|Δ| 16.23 | 129 claims · Σ|Δ| 17.27 | 127 claims · Σ|Δ| 17.78 | 125 claims · Σ|Δ| 16.13 |
AI_PAUSE_2028 prior 10% | 127 claims · Σ|Δ| 16.29 | 129 claims · Σ|Δ| 17.27 | 127 claims · Σ|Δ| 17.78 | 126 claims · Σ|Δ| 16.19 |
NO_AI_PAUSE_5Y prior 75% | 124 claims · Σ|Δ| 15.41 | 131 claims · Σ|Δ| 16.45 | 129 claims · Σ|Δ| 17.02 | 123 claims · Σ|Δ| 15.29 |
Method note
Joint conditional probability is approximated via log-odds combination: logit(P(pred|A,B)) ≈ logit(P(pred|A)) + logit(P(pred|B)) − logit(P(pred)). This is the closed-form Bayesian update assuming A and B are conditionally independent given the prediction. It's correct when the two scenarios act on the prediction through different causal paths; it's pessimistic when they overlap. The exact joint requires running the Gibbs sampler with both scenarios clamped, which would be N×M=16 sampling runs (~12 minutes per refresh) instead of N+M=8 — a 2× cost for higher fidelity.