Field conservation’s two stacks

Before AI-assisted platforms, the standard method for processing camera-trap images was entering species, location, and time into a spreadsheet image by image: slow, error-prone, and inconsistent enough between projects to hinder collaboration. AI classification tools did not converge that world; they multiplied options. Researchers today choose among Conservation AI, MegaDetector, MLWIC2, Wildlife Insights, Camelot, Timelapse, and others for overlapping slices of the same problem.

The sharper point for your public work: those tools stop at classification and research data management. They do not become a portfolio grid, an Impact Map pin, or a photo detail page with taxon context. Outside ArtInStack you still re-explain the same sighting in a second system.

On ArtInStack the science spine is Survey → Deployment → Media → Observation. The story context is Portfolio; Impact Map, page view, and SEO/Taxon are how that portfolio presents publicly when you opt in: siblings under Portfolio, not a second pipeline. Two contexts on purpose: field structure for partners and export, curation and privacy controls for visitors.

They meet on the media row. An observation tags science identity (and optional survey/deployment) onto a file; that same file can live in a portfolio for the public path. Same asset, two contexts; not a missing observation between disconnected apps.

We are not competing with Wildlife Insights or MegaDetector’s classification models. The defensible claim is narrower: once an observation exists on media, ArtInStack is where that file can carry survey structure and public story without a second upload into a website builder.

You bring classified sightings in; we keep Survey and Portfolio linkable through media and observation, without forcing wedding or commercial tenants to see field UI they never asked for.

Next: why observations speak Darwin Core / Camtrap instead of homemade columns. See Why open standards (Darwin Core). For the highest-stakes publishing judgment (coordinates, IUCN, and what the public panel should hide), read Publishing sensitive species. For org settings that enforce those choices across Evidence, maps, and exports, see Data Governance for conservation. For composing public Evidence under donor skepticism, see Evidence storytelling for skeptical donors. For DOI-backed publishing under an endorsed institution, see Publishing to GBIF: endorsement first. For click-path how-tos (surveys, tagging, export, toggles), use Conservation and fieldwork docs.