New modeling suggests Earth has about 20 million insect species, not 7 million

Earth may be home to about 20 million insect species, according to a new set of calculations reported by Scientific American. The headline number is the whole point: it lands at roughly three times more insect diversity than researchers previously thought.

So what changed, exactly? The reporting points to “some creative calculations” that stitch together evidence from bug traps, epidemiology, and trees. That combination matters because insect species are notoriously hard to count directly. You cannot simply send a satellite for the taxonomy layer, and you cannot measure every habitat exhaustively. Instead, scientists use proxy data from many angles, then infer the unseen. In this case, bug traps offer field sampling signals, epidemiology helps connect biodiversity patterns to how organisms interact in real-world biological systems, and “trees” refers to evolutionary relationships that can extend what we see to what likely exists.

If you are an executive or board member, it is tempting to file insect diversity under “environmental science, not corporate governance.” But the second-order implications travel fast. Insects are entangled with agriculture, forestry, supply chains, and public health, and they sit at the base of many food webs. When biodiversity estimates change by multiples, it can ripple into how organizations model ecological stability, pest pressure, and resilience. A threefold undercount is not a rounding error. It is a structural change in your mental model.

There is also an incentives angle that rarely shows up in headlines. Conservation funding and regulatory frameworks depend on what is considered “known.” Many compliance systems and environmental reporting regimes lean on documented species counts, habitat assessments, and risk frameworks that are partly shaped by the best available scientific estimates at the time. When a new synthesis suggests that the baseline is much larger than before, the practical burden is that regulators and planners may need to revisit categories of vulnerability, priority geographies, and the intensity of monitoring. Even if regulators do not immediately rewrite every rule, they can tighten scrutiny on how organizations justify environmental impacts and management plans.

This is where the methodology becomes more than academic trivia. The use of bug traps emphasizes that sampling can be biased by where you look and how long you look. Epidemiology points to how insects and related organisms show up in disease and ecological processes, which can surface effects that pure field trapping misses. And evolutionary “trees” anchor the inference: if you understand how insect lineages diversify over time, you can project diversity beyond observed samples. The point is not that any single proxy is perfect. The point is that triangulation can pull the estimate in a different direction, especially for groups that are hard to survey comprehensively.

Now connect that to how boards think. Risk management is about distributions, not point predictions. If the true distribution of species diversity is broader than assumed, then the tail risk associated with ecological disruption, invasive species dynamics, and ecosystem services can be larger than expected. That is relevant well beyond “nature lovers.” Agricultural companies, insurers, logistics operators, and any firm exposed to land use change all rely on assumptions about how ecosystems behave. Bigger biodiversity can mean more resilience in some cases. It can also mean more interactions and more pathways for pests, parasites, and ecosystem shifts. Either way, uncertainty can widen, which changes how much confidence you should place in older baselines.

Finally, zoom out to strategic timing. Scientific American’s report frames the estimate as the result of “creative calculations,” which signals an active frontier, not a settled final number. For decision-makers, the takeaway is not to treat 20 million as a precise census. The takeaway is to recognize that biodiversity accounting is still improving, and that current planning might be built on older underestimates. When the foundational counts move, the downstream frameworks that depend on them may need adjustment. For executives who oversee sustainability, risk, or compliance, the smart move is to treat these shifts as material signals about model quality, monitoring rigor, and the limits of what your stakeholders think is “already measured.”

In short: Earth might have about 20 million insect species, not far fewer. And when the known world expands by a factor of three, every organization operating in or beside ecosystems should expect more pressure to prove it understands what it is managing.