Invertebrates Are 75% of Known Species and 18% of Screen Time. We Are Part of the Problem.

In March 2023, People and Nature published a study by Kate Howlett and colleagues at the University of Cambridge: Wildlife documentaries present a diverse, but biased, portrayal of the natural world.

They compiled 945 nature documentaries produced between 1918 and June 2021, split them into seven time periods, and sampled 15 at random from each — 105 in total. For every one, they recorded each habitat, organism and species featured, and whether a conservation message appeared.

Then they compared the screen to the world.

The number

Birds and mammals alone account for more than half of all mentions. Plants are consistently low across every era.

Vertebrates are overrepresented by roughly 24 times. Nearly everything alive is an invertebrate, and nearly everything on screen is not.

The subtler finding

Even when invertebrates appear, they appear differently.

41.8% of vertebrate mentions were identifiable to species. For invertebrates it was 7.5%; for plants, 10%.

A vertebrate gets a name. An invertebrate gets a category.

The documentary says "a snow leopard" and "insects." One is an animal with a life; the other is set dressing — a mass noun, like weather. You cannot care about a mass noun, and you certainly cannot fund its conservation.

Why this is not just an aesthetic complaint

Howlett is direct about the cause, and fair about it: if you want people engaged, you need animals they already recognise and care about, or they will not watch. That is not cynicism. It is the constraint every wildlife filmmaker actually operates under.

Her caution is the point — the bias risks leaving people with the impression that "all of nature is fine."

But there is a harder finding underneath. The study notes its results are consistent with recent work showing conservation science itself has focused more on vertebrates over the last three decades.

Follow that chain:

What gets filmed → what the public cares about → what gets funded → what gets researched → what we know → what we can protect.

The camera is upstream of the science. A bias in what is filmed becomes a bias in what is studied, which becomes a bias in what can be conserved, because you cannot protect a species nobody has assessed.

This is why the IUCN's recent addition of more than 1,000 fungi species to the Red List matters more than it sounds. An entire kingdom was, until recently, almost entirely unassessed. Nobody filmed it either.

The habitat finding nobody noticed

Buried in the same study: the most-mentioned habitat is tropical forest. The least-mentioned is the deep ocean. This did not change significantly across a century.

Now read that against our report on deep-sea mining: 31 exploration contracts issued, a commercial permit expected by early 2027, 40 nations seeking a moratorium, and almost every species down there undescribed.

The least-filmed habitat on Earth is the one about to be mined.

That is not a coincidence — it is the mechanism. There is no constituency for a place nobody has seen. The abyssal plain has no David Attenborough episode, and so it has no public, and so the decision gets made without one.

The genuinely good news

The study found real, measurable improvement, and it would be dishonest to bury it:

• Conservation was mentioned in 16% of documentaries overall — but in almost half in the current decade, reaching 47% in the 2020s.
• No documentary before the 1980s contained a conservation message at all.
• Anthropogenic impacts appeared in 22% of documentaries, and never before the 1970s — with overexploitation of animal populations the most-cited.

For most of the history of the form, nature documentaries showed a pristine world with the humans edited out. That has changed within living memory, and it is changing fast.

The medium is getting more honest. It is not getting more representative.

Now the part about us

AEDO1 is being built as a documentary platform. We have not launched, we are not selling anything, and we have no library. So this study is not yet a finding about us — it is a finding about the industry we are walking into, and we would rather write it down now than discover it later.

Any platform that pays creators a share of what viewers pay will reward whatever viewers choose to watch. That is not a flaw in a particular business model; it is what a revenue share is.

And across a century, viewers choose mammals and birds.

So a creator who spends two years on the invertebrates of a single watershed will earn less than one who films a snow leopard — on our platform or anyone's. The arithmetic will be honest. The input will be skewed. Naming that before we build the thing is the only moment at which naming it costs us nothing, which is precisely why it is worth doing now rather than in a launch post.

We are not going to promise virtue we have not built. What we can say today:

• We will not claim a platform corrects a bias that a pay-per-view structure amplifies.
• The Conservation Report is free, is not paid per view, and carries no advertising — which means it can cover things nobody would pay to watch. This article is an example. Nothing about publishing it helps us commercially.
• If we ever build curation or funding that deliberately offsets this, it will be because there is a mechanism — not because we said we cared.

Howlett's own suggestion is the practical one, and it costs nothing: filmmakers have scope to focus on urban wildlife, showing people the biodiversity in their own area. The invertebrates are outside, now, where the viewer lives.

Why this is on AEDO1

We have spent this Report examining other people's numbers — the counterfactuals in carbon credits, the models behind Indigenous land projections, the gap between declared and effective ocean protection.

This is the one aimed at the industry we are joining. The finding is that it has spent a century teaching the public that nature is mostly vertebrates, and that conservation science followed the camera.

We will inherit that the day we launch, and a pay-per-view model will reinforce it. Publishing the study that says so does not fix anything — but a platform unwilling to print the research that indicts the economics it is about to adopt has no business calling itself a source of verified information.