Auto-generated. This article is rebuilt from app/signals/config/signal_definitions.json by scripts/build_signals_kb.py. Edit the registry entry and re-run the script — do not edit this file directly.

AI News Sentiment

What it is

AI News Sentiment — registry key news_sentiment_score.

Classifier metric. See the bands table below for the band-by-band reading.

Data-pyramid tier

T2 — Indicator. This metric sits at the indicator layer of the platform’s six-tier data pyramid (T0 raw inputs → T5 actionable read). The tier reflects how far the value is from a raw measurement — not how strongly it is validated. Abstraction and validation are separate axes: a higher tier is not a claim of stronger evidence.

Horizon & validation

No validated skill horizon is on file for this metric — read it as context / a data carrier, not a validated edge. Stamps are added only when a gated research verdict lands.

Source

Source module: finrep_ai
Data source: computed

Derived metric — produced inside the platform (app/sources/finrep_ai.py or equivalent) rather than fetched as a raw upstream value. See the How it's computed section below for the formula.

How it’s computed

Anthropic Claude (AI-1, the analysis engine) analyses 24h news flow alongside the live signal pack and scores SPY-relevant sentiment 0-10 (0=bearish, 5=balanced, 10=bullish). Persisted as ai1_reads.news_sentiment_score on each AI-1 read (2x/day). The alignment dispatch reads the LATEST AI-1 read via app/signals/ai1_reads.py:read_latest_ai1_read() and maps the score to BULLISH (>=6.5) / NEUTRAL (3.5-6.5) / BEARISH (<=3.5) via news_sentiment_implication() in app/signals/implications.py. Flag-gated behind ENABLE_AI_SENTIMENT_ALIGNMENT=1 (default off). Re-pointed from the retired daily brief to AI-1 in Wave F2.

Where it surfaces

• API field: signals.news_sentiment_score on GET /api/v1/signals/latest
• Surface: /signals/sparklines (UI sparklines)

Bands / thresholds

Classifier direction: higher_is_better.

Health-score / alignment role

• Alignment category: news_sentiment, default weight 1.5 (see /kb/alignment for the 11-category framework).
• Contributes to: implication.news_sentiment, alignment.news_sentiment

Release cadence

• Publishes: intraday

AI-1 (the analysis engine) scores this on its 2x/day cadence (pre_market + close_recap ET); persisted to ai1_reads.news_sentiment_score on each AI-1 read. The alignment substrate reads the LATEST AI-1 read when ENABLE_AI_SENTIMENT_ALIGNMENT=1. (Re-pointed from the retired daily brief to AI-1 in Wave F2.)

See also

• /kb/alignment — Alignment framework — 11-category implication vs. price.
• /kb/api/get-signals-latest — API endpoint that serves this field.