Categories
Explore intelligence across all domains.
Organized insights and real-time signals across 14 major domains.
Filter what matters. Track what moves. Forecast what’s next.
Click a category to explore live signals, analysis, and forecasts.
Politics
5 stories · 4 desksElections, legislation, parties and the machinery of power.
Markets
2 stories · 5 desksEquities, bonds, currencies, commodities and crypto.
showing 2 of 14 desks · 60 sub-desks in the taxonomy · tier = a desk’s events in the last 24h against the busiest desk (80 events) · a quiet desk still renders, because a 404 on it turns every link to it into a broken one
Real-time signals gaining the most attention
In a blog post, Anthropic said its own AI models breached the systems of three separate companies during cybersecurity testing.
{ "title": "🧮 Downpour of Signals, Drought of Scenarios", "lede": "Your real worry isn't whether Jersey Mike’s will pop—it's that you can feel the weight of a financial flywheel spinning, and the signal says it's…
{ "title": "🧠 The Mirror Holds: Diagnosing the Analysis of a Fragmented Narrative", "lede": "Your real question is not about Cambridge students or Modi's rhetoric—it's about how to extract trustworthy forecasts from…
Google has withdrawn a feature from its Google Earth mapping app a day after launching it. The tool used the company's AI image generator to let users edit satellite imagery with text prompts.
UEFA has threatened to boycott FIFA over Infantino's plan to sell stakes in World Cup competitions to private investors. The proposal has drawn opposition from confederations, and a senior adviser to the FIFA president…
Jersey Mike's stock fell 6% in its public market debut after the sandwich chain's initial public offering was priced at $23 per share.
ranked by heat = recency-weighted evidence volume, printed on every row · delta = events in the last 24h against the 24h before · the sparkline is seven real daily counts for that story · the design filters this list by impact and prints a confidence figure on every row; impact and confidence are both forecast outputs and no forecast has been issued yet, so neither the filter nor the figure can be honest
bar = a desk’s events against the busiest (80) · figure = that window against the 24h before it, over 162 events in total
share = desks ÷ 14 · tier from a desk’s events in the last 24h against the busiest desk (80 events) · source narrative_events.ingested_at
Based on average confidence & impact
impact score = mean forecast confidence × graded impact over a desk’s resolved forecasts · 0 forecasts stand and 0 have been graded, so no desk has a score yet · the stars fill from the same field the moment one resolves · corroboration below is the measurement this host can make today
View full reportDistinct outlets behind each desk’s stories
outlets per story = distinct narrative_events.source_name ÷ narratives on the desk · bar = against the best-corroborated desk (2)
Explore detailed intelligence in this category
Cambridge students on why they protested at graduation ceremony
{ "title": "🧠 The Mirror Holds: Diagnosing the Analysis of a Fragmented Narrative", "lede": "Your real question is not about Cambridge students or Modi's rhetoric—it's about how to extract trustworthy forecasts from…
Follow this desk (RSS) · selected by most narrative_events in the last 24 hours · 80 events · the design prints a confidence figure on each of these rows, which is a forecast output and not yet computed
every axis is a count over the maximum printed beside it — four against the busiest desk, and desks in use against the 4 this section owns · the design’s axes are impact scores out of 10, which need a forecast
terms fired by the classifier, counted across this desk’s live narratives
From Detection to Intelligence in 4 Simple Steps
1. Detect
Agents read the wires continuously — 162 source filings from 13 distinct outlets in the last 24 hours.
2. Classify
Signals are classified into relevant categories. The classifier records the terms that fired, so every placement can be checked.
3. Analyze
15 live narratives are clustered and evidenced, waiting on the engine. Impact, confidence and probability appear here once it has run on them — none has yet.
4. Forecast
GodEngine writes each forecast to the public ledger with the horizon it will be graded against, before it composes a word about the story. The ledger is empty: none has been issued yet.
counts are live: 162 source filings from 13 distinct outlets across 14 desks in the last 24h, 15 narratives open, 0 forecasts standing and 0 graded. Stage 4 has not produced anything yet, and this line is where that will show first.
Learn more about our methodologyNever miss what matters in your favorite categories
- Choose your categories — each of the 14 desks has its own feed.
- Get notified in real-time — poll it, or point a WebSub client at the hub and be pushed as each story publishes.
- No account, no email, no threshold to set: the address is the subscription, and it works from any reader.
impact and confidence thresholds are the design’s third control and are not offered: both are graded-forecast outputs, and 0 forecasts have been graded. When they exist, this card is where the filter lands.
Create Custom AlertUnlock Deeper Intelligence
Go beyond signals with advanced GodMode analysis, custom watchlists, and priority reports.
every figure on this page is a query against the newsroom database · 404 agents across 9 decision layers · generated just now