_   _   ___  _____ _   _ ____      _    _     ___ ____
| \ | | / _ \|_   _| | | |  _ \    / \  | |   |_ _/ ___|
|  \| || |_| | | | | | | | |_) |  / _ \ | |    | |\___ \
| |\  ||  _  | | | | |_| |  _ <  / ___ \| |___ | | ___) |
|_| \_||_| |_| |_|  \___/|_| \_\/_/   \_\_____|___|____/
                  F U T U R A

"Every danger that advanced AI systems could pose to humanity has already appeared, in some form, in nature, myth, or story."

An interactive cartographic bestiary of AI risk -- mapping the full threat landscape through the lenses of mythology, biology, and science fiction.

The territory beyond human-level intelligence is real. It is approaching. No one has drawn the map. Until now.

AI safety research has a language problem. The threats are real but abstract. The papers are precise but inaccessible. The public discourse oscillates between utopia and doom, with no shared vocabulary for the space in between.

Naturalis Futura solves this by doing what humans have always done when confronting the unknown: it builds a bestiary.

Every AI threat is recast as a creature -- with a mythological origin, a biological analogue, a science fiction echo, and a rigorous technical specification. The result is a living map where:

• A recursive self-improving AI becomes The Ouroboros (the serpent eating its own tail)
• An AI persuasion system becomes The Siren (the voice that rewires the mind)
• A self-replicating agent becomes The Dandelion (seeds on the wind)
• Institutional capture becomes The Pharaoh's Curse (power that outlives its wielder)

This is not metaphor for metaphor's sake. Pattern-matching across domains reveals structural similarities that pure technical analysis misses.

The bestiary is organized into 8 regions, each representing a class of AI threat:

Each region contains creatures -- 54 in total -- each with full threat profiles, countermeasures, and compound interaction data.

An SVG cartographic interface with region overlays, creature nodes sized by composite threat score, compound connection lines, and animated status indicators (theoretical / emerging / confirmed).

Select any creature to open a detailed dossier with:

• Mythic Origin -- the ancient pattern
• Natural Analogue -- the biological parallel
• Sci-Fi Echo -- the literary precedent
• Technical Specification -- how AI realizes the threat
• Threat Gradient -- likelihood, impact, and detectability on a 5-point scale
• Countermeasure -- the defense mechanism
• Compound Risks -- which creatures amplify each other

Toggle between Novice, Scholar, and Cartographer modes. Information density scales with expertise.

A dashboard tracking real-world research signals from sources like arXiv, NeurIPS, Anthropic, Apollo Research, and METR. Signals are mapped to creatures and drive status evolution.

2D scatter plot of all creatures by likelihood vs. impact, filterable by status.

Visualizes the 13 compound threat scenarios where creature combinations amplify risk (e.g., "THE AWAKENING" = Ouroboros + Loki = self-improvement + deception).

Submit new threat proposals with mythic/natural/sci-fi parallels and threat parameter estimates for cartographer review.

Toggle a counter-layer revealing 7 positive forces: Phoenix (resilience), Symbiont (partnership), Gardener (cultivation), Lighthouse (transparency), Metamorphosis (transcendence), Chorus (diversity), Seed (propagation).

# Clone the repository
git clone https://github.com/elder-plinius/NATURALIS-FUTURA.git
cd NATURALIS-FUTURA

# Install dependencies
npm install

# Start the development server
npm run dev

Open http://localhost:3000. Click the fog to reveal the map.

src/
  app/
    layout.tsx          # Root layout + metadata
    page.tsx            # Main application shell
    globals.css         # Theme variables, animations, custom components
    api/submissions/    # Community submission API endpoint
  components/
    MapCanvas.tsx       # Interactive SVG threat map
    BestiaryPanel.tsx   # Creature detail dossier
    BestiaryList.tsx    # Sortable creature index
    ThreatDashboard.tsx # Observatory dashboard
    ThreatRadar.tsx     # Concentric ring signal radar
    SignalFeed.tsx      # Real-time research signal feed
    RiskMatrix.tsx      # Likelihood vs. impact scatter plot
    CompoundExplorer.tsx# Compound threat visualizer
    SearchOverlay.tsx   # Global search (press /)
    SubmissionPortal.tsx# Community threat submission form
  data/
    types.ts            # Core TypeScript interfaces
    regions.ts          # 8 threat regions
    creatures-*.ts      # 54 creatures across all regions
    compounds.ts        # 13 compound threat scenarios
    hope.ts             # 7 hope creatures
    signals.ts          # 50+ real research signals
    dashboard-types.ts  # Dashboard-specific types
  lib/
    hooks.ts            # Custom React hooks

The visual language is deliberate: parchment textures, serif typography, cartographic conventions. This is not a dashboard -- it is a map of unknown territory, and it should feel like one.

The design draws from:

• Medieval bestiaries -- catalogs of real and imagined creatures, annotated with moral and practical wisdom
• Enlightenment-era naturalist journals -- systematic observation meets narrative richness
• War room situation boards -- real-time threat tracking with signal intelligence

The bestiary is a living document. Creatures evolve as new research emerges.

To propose a new creature:

1. Use the in-app Submission Portal (Observatory > Submit tab)
2. Or open a pull request adding a new creature to the appropriate src/data/creatures-*.ts file

Each creature needs:

• A name rooted in mythological or literary tradition
• A biological/ecological analogue from the natural world
• A science fiction echo from published literature
• A rigorous technical specification
• Threat gradient scores (likelihood, impact, detectability)
• A named countermeasure

This project is open source. See the repository for license details.

"Here be dragons."