How AI technology is aiding Mountain Bongo conservation in Kenya

Scientists at the Mount Kenya Wildlife Conservancy are lauding the introduction of artificial intelligence-driven camera technology in their Mountain Bongo conservation, breeding, and rewilding program for significantly enhancing efficiency and success in re-establishing the critically endangered species in its natural habitat.

According to researchers, the application of this technology—a combination of AI-enabled cameras, software, and several mobile and computer applications—now places Kenya at the cutting edge of global conservation.

"Last year, we began pioneering the use of artificial intelligence to track and monitor these bongos. We first rolled out the technology in our sanctuary to pilot it and to enable machine learning for the camera—to collect enough datasets for it to recognize the bongo. The camera needs to view at least 5,000 images of the animal to recognize it efficiently," says Dr. Robert Aruho, Team Lead at the Mount Kenya Wildlife Conservancy.

"The AI technology has worked so well that we are now able to deploy more cameras in our animal sanctuary and enroll that same in the Aberdares and they have given us very good results," he adds.

The conservancy, which sits on approximately 1,250 acres, comprises an animal orphanage for injured animals that would not survive in the wild, and a 776-acre sanctuary where animals roam free in a vast, forested enclosure adjoining Mount Kenya National Park. Animals that thrive in the sanctuary can later be released into the Aberdares forest.

• .

  Keep Reading

• 
  AI-powered humanoids likely to take over factories by 2027, Chinese firm says
• 
  Broad-based bait: Kenya Kwanza on expedition to fish Kalonzo
• 
  17 mountain bongos make epic return to Kenya from the US
• 
  Artificial Intelligence can be used for good, guard against its dangers

This conservation effort dates back several decades. In 1964, conservationists exported 36 Mountain Bongos to the United States to preserve the dwindling population, with the hope that their descendants could one day help repopulate the species.

Today, fewer than 100 individuals are left in the wild. According to Dr. Aruho, these individuals are too geographically scattered to serve as viable breeding partners.

"The Mountain Bongo population in Kenya has been falling, with smaller groups remaining in the Aberdare Ranges, Eburu, and the Mau Forest. These populations are severely separated, making breeding and natural recovery nearly impossible. So the only way to save them is through captive breeding and rewilding."

In 2004, the conservancy—which began decades earlier as the Mt. Kenya Game Ranch—launched its  Mountain Bongo Conservation Program.

“We imported 18 bongos from the U.S., and since then, we’ve been breeding them and preparing them for survival in the wild. So far, we’ve produced four generations, and in 2022, we released the first batch into the sanctuary. In 2024, we released another, and the results have been quite promising,” Dr. Aruho says. “We are now able to observe the animals in their natural habitat without interrupting their behavior, allowing them to truly be bongos.”

For years, scientists tracked Mountain Bongos either on foot or by manually placing range cameras in areas where they suspected the animals would feed or pass. This approach was not only physically demanding but also intrusive, leaving behind human scent that altered the animals' behavior.

“This new technology has helped us better understand the adaptive behavior of the bongos we are rewilding, and compare it with the behavior of those in the wild,” Dr. Aruho explains. “It has also reduced our manpower needs. We used to spend six hours searching for bongos in the bush—now, we can spot them on camera before even arriving at the sanctuary.”

Dr. Aruho notes that remote monitoring has brought other unexpected benefits as well.

“We can now monitor these animals continuously, day and night. The technology gives us real-time data even from areas inaccessible to humans—like dense forests or large swamps. That drastically reduces logistics while providing valuable insights.”

“It also enhances security. If there’s a poaching threat, we now receive alerts in advance. And for human-wildlife conflict management, we can spot an animal that’s likely to stray and act preemptively. That, to me, is a game changer,” he adds.

Such progress would not have been possible without key partnerships.

Collaboration with the Kenya Wildlife Service has been essential, allowing the team to install cameras within Aberdare National Park and extend their monitoring of the bongos’ natural habitat.

Individual philanthropy has also played a critical role.

In 2014, the conservancy’s major benefactor, Humphrey Kariuki, came on board. The Mount Kenya Wildlife Conservancy is funded by the Janus Continental Group (JCG), founded by Kariuki. Staff at the conservancy say he often recalls his youth, when he would chase Mountain Bongos from his mother’s kitchen garden in what is now Nyeri County.

“He remembers fondly the times he had to chase the bongos from his mother’s crops. Today, there are none left in Nyeri,” a staff member shared during a briefing.

The conservancy also receives support from other donors and well-wishers, contributing up to 10% of the annual budget needed for its operations.

Technological partnerships have been equally crucial.

The application of AI and machine learning is underpinned by research from Chester Zoo in Cheshire and Liverpool John Moores University in the U.K. Jane Mwihaki Wathika, a research scientist, explains how AI-powered cameras have become indispensable.

“Under our partnership with Chester Zoo, they donated around 15 AI cameras, some of which have been mounted at the Mawingu Sanctuary and in Aberdare National Park. We identified key areas—like elephant corridors, feeding zones, and watering points—where the cameras would be most effective,” she says.

“The cameras are linked to an app, allowing us to monitor them in real-time from the field. Previously, we had to travel to retrieve memory cards.”

Another benefit: the cameras are solar-powered, minimizing the risk of downtime due to power issues.

However, the project has not been without its challenges. Equipment has been destroyed by animals, and many of the target areas are remote, with poor or no mobile network coverage.

Researchers have tried engaging mobile service providers to improve coverage, with little success. Dr. Aruho describes network access as one of the biggest hurdles.

“It’s a real challenge. We keep switching between Safaricom and Airtel depending on where the signal is stronger. But that forces us to deploy based on connectivity—not where bongos are likely to be. Because of this limitation, we might be missing something.”

“We’re now exploring satellite internet as a solution. This remains a top priority if we are to scale this technology across the entire bongo range.”

It’s ironic that such a basic technological barrier is slowing down a project that is otherwise employing cutting-edge AI. But the reality is that mobile service providers invest in coverage where there’s a viable market, not in the deep, forested, mountainous regions the bongos call home.

With the success of the rewilding program and a thriving fourth generation of reintroduced Mountain Bongos, there is now real hope that this shy, chocolate-brown antelope may once again roam freely in the wilds of central Kenya.