The Red List in 2019 identified that out of 195,000 species, 28,338 are nearing extinction, all thanks to human activity, and today, the count has reached 32,000, including mammals and birds. Jan Borgelt, an ecologist at the Norwegian University of Science And Technology and the lead author, says, “Things could be worse than we realize now,” “More species are likely to be threatened than we previously thought.”
Though the damage is caused by humans majorly, a little bit of nature is also to be blamed for the current situation. However, there are so many species on earth that we are unaware they exist, but we know they are nearing extinction. Everything about how artificial intelligence is used to create devices in the century’s political backgrounds.
A new study that was carried out using machine learning to understand how serious the situation is shows a problematic result. Some plant and animal species did not have enough data, so they were labeled data-deficient. As a result, the conservationists could not collect enough information to understand their lifestyle or how many were left.
The species with “data deficiency” are even more threatened than those with data. The data from the study conducted by the International Union for Conservation Of Nature maintains a list globally called Red List; surveys rank the animals on how threatened they are. 56% of “data deficient” species are at risk of extinction, and only 28% of the species whose whereabouts are understood. Moreover, decoding the animals’ language via using AI.
Borgelt’s work focuses on how human activity affects biodiversity and the ecosystem through plastic pollution and hydroelectricity. Though the Red List is a great resource, it lists almost 20,000 data deficient species doing the research relying on this far from precise.
To solve the above problem, Borgelt and his team took help from machine learning. First, they trained algorithms for understanding the data deficient species’ extinction risk. For this purpose, they took help from the IUCN’s data on 28,363 animals that had been evaluated. As a result, algorithms could understand the factors responsible for how threatened a species was, which included invasive species, pollution, and climate change.
After this, the team turned towards the data deficient species of around 7,699, but they could only work with the species whose geographical distribution of animals was known. The algorithm showed that 56% of those species were at risk of extinction, and some animals are more likely to become extinct than others, around 85% of amphibians. The amphibians included spotted narrow-mouthed frog, Mali Screeching frog, and rubber frogs. It would be interesting to see these frogs, but IUCN does not even have its photos on the Red List.
The research was validated a little when the Red List was updated last year. 76% of the algorithm’s predictions were correct. One hundred twenty-three species in the updated Red List were added, which was predicted by the algorithm. European legislators are moving toward passing an ‘Artificial Intelligence Act’ for the region to protect the public against its adverse effects.
Borgelt said, “That was reassuring,” Also, he stated that he understands the limitations of machine learning. Further added, “For now, [these algorithms] certainly shouldn’t replace expert assessments,” and said that the assessments from the experts are more accurate. Borgelt feels, “But such algorithms, they’re swift. They’re not so time intensive or labor intensive as if you were to assess the species individually,”
There are plenty of reasons for the species numbers to have been looked away by the researchers. Like the killer whale that made its way in the ’90s movie comes in the data deficient species list. Scientists are unaware that the killer whale is one of its kind or that there are more. In addition, some animals are found only in some regions, making them challenging to study and vulnerable. This particular reason makes it more important to learn about these animals.
Borgelt reiterates that machine learning “isn’t a replacement for tracing down the animals on the ground. But it’s another tool in the toolbox that could help conservationists figure out which species need some extra TLC.”