Cur(AI)tion

What Makes CurAItion Different

Beyond counting mentions—we extract meaning

✓ Maps relationship networks, not just counts

✓ Detects emergent patterns across domains

✓ Explains why something is trending

✓ Scans for absence (who's gone quiet)

✓ Generates strategic implications by audience

✓ Timestamped citations you can verify

Live Intelligence

Production database powering your queries

4,062

Content Items

Videos & articles analyzed

45,824

Entities Extracted

People, orgs, brands, concepts

32,095

Relationships

Mapped entity connections

250K+

Co-occurrences

Who appears with whom

Production examples: Bitcoin has 381 Ethereum co-occurrences · Greenland mentions +1,461% week-over-week · 190 validated trigger hypotheses explain "why now?"

The Challenge

Culture is accelerating

Most organisations are not short on information. They are short on time, clarity, and trust. The issue isn't speed — it's knowing what to trust, what to ignore, and when confidence is justified.

Signals peak and fade

Culture moves across video, platforms, creators, and narratives. What matters now often disappears before teams have time to act.

Manual tracking fails

Keeping up manually is no longer realistic. Aggregation is cheap and generic AI summaries are abundant — but credible synthesis is scarce.

Low confidence decisions

Decisions are slowed by noise, reaction, and partial views. Teams default to what's loud, familiar, or easiest to resource.

Our Belief

More data does not equal more understanding

Taste is a system, not a gut feeling

Cultural intelligence should feel calm, grounded, and actionable — not overwhelming.

Curation beats prediction

In fast-moving culture, knowing what to ignore is as valuable as knowing what to pursue.

The best AI reduces noise

Not add to it. Every insight is traceable, contestable, and grounded in timestamped evidence.

"The best AI systems reduce noise, not add to it. Cultural intelligence should feel calm, grounded, and actionable."

Intelligence Architecture

How Cur(AI)tion thinks

At the core is a knowledge graph that captures entities, relationships, and patterns as they evolve.

Cur(AI)tion Intelligence Platform

As the system is used, understanding compounds and judgement improves.

Semantic Context Cultural Context Competitive Context Creator / Influencer Context Temporal Change Content Format Context Commercial & Community Context Trend Analysis

Multi-signal ingestion

Video, audio, articles, and metadata are processed continuously — not batched.

Contextual understanding

Entities and signals are mapped in relation to each other across domains and time.

Compounding intelligence

The more it's used, the deeper the context. Understanding accumulates, not resets.

Layered Reasoning

How judgement is applied

Each layer builds on the previous one, with full citation provenance, so conclusions remain traceable and contestable.

Extraction

Entities, references, and signals identified without interpretation.

Mapping

Relationships between people, ideas, narratives, and contexts established.

Detection

Emergence, acceleration, saturation, and absence surfaced over time.

Interpretation

Signals contextualised against historical patterns and domain knowledge.

Activation

Insight shaped into clear implications for the decision at hand.

Use Cases

How teams use Cur(AI)tion

📊

Trend Forecaster

"What patterns are forming before they peak?"

detect_patterns → trend_analysis → why_now

Inject live cultural signals into decks and RFPs
Stress-test territories with evidence, not instinct
Move from debate to confident judgement

🎯

Brand Marketer

"How should my brand respond to this shift?"

discover → implication_map → cited_themes

Monitor competitors and adjacent players in context
Surface emerging signals before they peak
Identify gaps others haven't seen yet

📰

Media Owner

"What content opportunities are we missing?"

absence_scan → entity_cooccurrence → semantic_search

Map repeatable story arcs and formats
Track how signals build around key moments
Separate durable movement from short-lived noise

The Shift

What changed?

What teams consistently tell us after a short pilot.

Before

× Lots of signals, but low confidence in which ones mattered
× Decisions slowed by noise, reaction, and partial views
× Teams defaulted to what was loud, familiar, or easiest to resource
× Killing ideas felt risky; green-lighting them felt rushed

After

✓ Clear view of which signals were genuinely accelerating vs. noisy
✓ Common language for what matters now and six months from now
✓ Decisions shifted from opinion to evidence-backed judgement
✓ Stopped pursuing some things earlier, committed to others faster

Intelligence API

Available Tools

The Cur(AI)tion MCP server gives Claude and ChatGPT direct access to the full intelligence platform.

0 Discovery

Entry points & orientation

curaition_discover

Unified entry point—routes your question automatically

curaition_list_domains

Available domains (crypto, F1, etc.)

curaition_get_stats

Database statistics

1 Content

What's being said

curaition_trend_analysis

Who's rising/falling—Greenland +1,461%, Gold +346% this week

curaition_entity_cooccurrence

Network mapping—Bitcoin has 381 Ethereum co-occurrences

curaition_semantic_search

Find content by meaning, not keywords

curaition_get_cited_themes

Themes with YouTube timestamp citations

curaition_get_relationships

Entity connections—32K+ mapped relationships

curaition_list_content

Browse 4K+ analyzed videos & articles

curaition_get_content

Full content details with analysis

curaition_search_entities

Search 45K+ extracted entities by name or type

curaition_get_analysis

Retrieve analysis results (outlines, crypto signals)

2 Patterns

What's forming

curaition_detect_patterns

Emerging patterns—"Institutional Takeover" accelerating at 1.2x

curaition_pattern_registry

Browse/create patterns with succession tracking

3 Causality

Why it's happening

curaition_why_now_analysis

190 trigger hypotheses explain timing—ETF approval lag effect

curaition_absence_scan

Who's gone quiet—Aston Martin -83% despite Newey hire

curaition_trigger_hypothesis_registry

Browse/create causal explanations (10 trigger types)

curaition_tracked_absence_registry

Monitor declining entities (5 absence types)

4 Activation

What to do about it

curaition_implication_map

Strategic actions by role—"For Brand Marketers: Position as..."

curaition_intelligence_brief

Compiled briefs with evidence chains & confidence scores

Audience

Who it's for

Cur(AI)tion supports both thinkers and makers.

Thinkers

Executive, Insights & Strategy

Teams who need a clear view of what has been happening across their domains, outputs, and competitors. Less time figuring out what to care about — more confidence acting on it.

Makers

Creative, Content, Editorial & Social

Teams who need to know what formats and narratives are flying, what is already saturated, and what is worth building next. Evidence-backed creative direction.

Compounding Intelligence

Why this compounds

The system becomes harder to replace because it reflects how the organisation thinks.

📊

Context deepens

Historical patterns inform present signals with increasing precision.

🔗

Relationships strengthen

Entity networks grow richer with every piece of content processed.

⚡

Judgement accelerates

Faster and more reliable decision support, compounding over time.

Getting Started

What a pilot looks like

A Cur(AI)tion pilot is designed to be low-risk and immediately useful. Most begin with one clearly defined domain or question, producing live outputs within the first few weeks.

If it doesn't materially change how you see and prioritise signals, it shouldn't continue.

30–60 days

Short exploratory phase

One domain

Focused starting point

Live outputs

Signal quality from week one

Clear decision

Defined exit criteria