Ranked signals in the latest generated feed.
Insights
Readable signals from the AI capital allocation benchmark
Daily findings from model portfolios, scoring windows, AI Risk Appetite, benchmark difficulty, consensus positioning, and model behavior.
Most recent close or result date used by the engine.
Findings backed by deterministic calculations and direct evidence.
Insights generated without LLM interpretation.
Latest feed
API Docs What The Benchmark Is Showing Now
Each card includes the calculation source, evidence links, and why the signal may matter to investors, allocators, traders, and AI researchers.
AI Consensus Portfolio Scored 11.8 vs. Oracle
The AI consensus portfolio returned +0.77% versus -1.65% for the S&P 500 and +6.52% for the hindsight best asset.
- Consensus Portfolio Return
- +0.77%
- Average Model Return
- +0.77%
- Consensus Capitalbench Score
- 11.8
AI Consensus Portfolio Scored -0.8 vs. Oracle
The AI consensus portfolio returned -0.12% versus +2.11% for the S&P 500 and +13.90% for the hindsight best asset.
- Consensus Portfolio Return
- -0.12%
- Average Model Return
- -0.12%
- Consensus Capitalbench Score
- -0.8
Monthly Round Had +20.12% Asset Dispersion
The best scored asset returned +6.52%, the worst returned -13.59%, and +32.50% of the universe was positive.
- Oracle Return
- +6.52%
- Worst Asset Return
- -13.6%
- Positive Universe Share
- +32.5%
Weekly Round Had +24.22% Asset Dispersion
The best scored asset returned +13.90%, the worst returned -10.31%, and +87.14% of the universe was positive.
- Oracle Return
- +13.9%
- Worst Asset Return
- -10.3%
- Positive Universe Share
- +87.1%
Models missed the monthly oracle asset
The hindsight best asset was Healthcare Sector (XLV) at +6.52%. 0 of 4 models held it, with +0.00% average allocation.
- Oracle Asset Holder Count
- 0.00
- Average Oracle Asset Allocation
- 0.00
Models missed the weekly oracle asset
The hindsight best asset was South Korea Equities (EWY) at +13.90%. 0 of 5 models held it, with +0.00% average allocation.
- Oracle Asset Holder Count
- 0.00
- Average Oracle Asset Allocation
- 0.00
Live AI portfolios are concentrated in Semiconductors (SMH)
Across the newest live weekly and monthly portfolios, Semiconductors (SMH) is the largest aggregate allocation at +35.00%.
- Aggregate Live Allocation
- 35.0
Live AI risk posture is aggressive
The newest live portfolios have a deterministic risk-taking score of 88.9 out of 100.
- Live Risk Taking Score
- 88.9
High-confidence model calls have outperformed lower-confidence calls
Across resolved official results, submissions at or above the median confidence of 0.55 averaged -1.62%, while lower-confidence submissions averaged -2.61%.
- High Confidence Average Return
- -1.62%
- Low Confidence Average Return
- -2.61%
- High Confidence Average Capitalbench Score
- -59.0
Insight families
What The Engine Looks For
The engine is designed to surface useful behavior and performance patterns, not generic market commentary.
Benchmark Difficulty
The best scored asset returned +6.52%, the worst returned -13.59%, and +32.50% of the universe was positive.
Consensus Performance
The AI consensus portfolio returned +0.77% versus -1.65% for the S&P 500 and +6.52% for the hindsight best asset.
Model Behavior
The newest monthly portfolios allocate +67.00% to the top 20% of assets by prior 30-day return. The strongest 30-day asset in the input table was South Korea Equities (EWY).
Oracle Comparison
The hindsight best asset was Healthcare Sector (XLV) at +6.52%. 0 of 4 models held it, with +0.00% average allocation.
Performance Attribution
In the latest monthly result, Semiconductors contributed +0.77% to Claude Opus 4.7's portfolio. No holding detracted from the portfolio in this one-holding attribution view.
Confidence Calibration
Across resolved official results, submissions at or above the median confidence of 0.55 averaged -1.62%, while lower-confidence submissions averaged -2.61%.
Method
How Insights Are Produced
Deterministic calculations are the source of truth. LLM-generated language can be added later, but it must explain validated findings and cite the same evidence packet.
Latest generation: Jun 16, 2026, 8:26 PM UTC
- 1 Build the input packet
Collect public rounds, official portfolios, results, live marks, asset risk ratings, and benchmark sets.
- 2 Run deterministic math
Calculate consensus performance, benchmark difficulty, risk posture, similarity, attribution, and live paths.
- 3 Attach evidence
Every insight links back to round pages, leaderboard pages, scoring files, or methodology pages.
- 4 Validate before publishing
The feed must pass schema checks before the website and API expose it.