JOHN YANG: In movies and literature, a wolfús haunting howl can signify danger or untamed nature.

In real life, researchers in Yellowstone national park are analyzing those howls with cutting edge AI technology to better monitor and track wolves.

Matt Standal of PBS Montana explains 653.

MAN: The wolves that have crossed the river are howling.

MATT STANDAL (voice-over): Here in Yellowstone National Parkús Lamar Valley, wolves from one of the parkús nine packs have made a kill.

MAN: Letús go ahead and feel free to take a look through there.

Thatús the Junction Butte pack.

MATT STANDAL (voice-over): Itús the end of August, the peak of bison rutting, or mating season, and a time when wolves increasingly prey on bison.

Many become injured or weakened during the fierce competition for a mate, and the wolves take advantage.

Park wolf technician Jeremy Sundaraj is monitoring the pack as they feed and teaching curious tourists about wolves.

JEREMY SUNDERRAJ, Wolf Technician: So what weúre trying to do here is just kind of count how many there are, record their behavior.

If we can see, like what the carcass is, this is almost certainly a kill just based on how theyúre behaving around it.

And if weúre quiet, we can actually maybe hear them howling.

MATT STANDAL (voice-over): Their howls have become central to a new cutting edge conservation project, using artificial intelligence to decode sound recordings.

This development in the field of bioacoustics could redefine how wolves like these are monitored in the wild.

MAN: You got it.

MATT STANDAL (voice-over): Since 1995, when gray wolves were reintroduced into Yellowstone, park biologists have used airplanes to spot them, helicopters to track them, and dart guns to tag them so they can be fitted with radio and GPS collars.

MAN: And you can see most of the wolf talking is happening in the night.

MATT STANDAL (voice-over): Now Bioacoustics is offering a new, less invasive way to study them, using sound and advances in artificial intelligence to one day potentially decode wolf communication by matching their howls with specific behavior.

DAN STAHLER, Senior Wildlife Biologist: Not only can we hear them here and record their howling 24/7, 365 days a year, but we often can link behaviors of wolves by observing them when they are vocalizing.

What is the cause and effect of howling?

MATT STANDAL (voice-over): Dan Stahler is the senior wildlife biologist for Yellowstone National Park, a job that includes gathering the data from sound recorders like this one hidden in a tree near park headquarters.

DAN STAHLER: One more Jeremy.

MATT STANDAL (voice-over): Stahlerús team has been recording the barks, yips and howls of Yellowstoneús nine wolf packs, more than 100 wolves, for the past year.

DAN STAHLER: Thatús another goal of ours, is can we detect unique pack signatures and use that?

MATT STANDAL (voice-over): He says, Theyúve collected over 7,000 wolf sounds and have been able to identify the acoustic signatures of several wolf packs in the park.

In the future, Stahler thinks this bioacoustic work could partially replace the hazardous duty of capturing and collaring wolves.

DAN STAHLER: And so what I could envision down the road a decade from now is that we may not have to collar certain packs or put collars out in certain areas of the park.

And then with new cutting edge AI tools we hope.

Weúre not sure yet, but we hope we can answer really interesting questions about what are wolves actually saying or can we count wolves?

Can we identify unique individuals?

JEFFREY REED: Good morning, everybody.

My name is.

MATT STANDAL (voice-over): Linguistics researcher and software engineer Dr.

Jeff Reed has been experimenting with AI to study wolves near his home just north of Yellowstone.

Heús lending his technical expertise to the Yellowstone Wolf Project.

JEFFREY REED: This is a wolf chorus howl and weúre using AI from Google to see if we can count the number of wolves a chorus howl.

So this is a group of wolves.

Itús like you walking into a bar and all the people are talking and you can pick out a particular person in the room.

Wolves can pick out other wolves that they know in this cacophony of sound.

MATT STANDAL (voice-over): The key to the technology is pattern recognition.

According to Reed, these colorful patterns are whatús called a spectrogram of wolf howls, representing their strength and frequency over time.

Artificial intelligence, he says, can pick out the patterns and identify individual wolves much faster than any human could.

JEFFREY REED: These battery operated devices use AI.

MATT STANDAL (voice-over): Reed leads a company that makes the high tech AI enabled field cameras and audio recorders that Yellowstone is using to monitor its vast space.

But these cameras, called Griz cams, donút just listen to wildlife.

They can also pick up human conversations and activities from hundreds of yards away.

Animal science and human privacy in Yellowstone could soon be on a collision course.

25 of these cameras will be installed in a grid across the park, thanks to a large donation from a company called Colossal Biosciences.

MATT JAMES, Colossal Biosciences: For me, the moonshot with bioacoustics and wolves is can we reduce the conflict between wolves and humans?

MATT STANDAL (voice-over): Matt James is the chief animal officer at Colossal Biosciences.

He says AI recording technology can be used to protect wolves from humans.

MATT JAMES: And can we explain that these are empathetic, emotionally complex animals that arenút mindless hunters and they deserve the ability to coexist with us.

MATT STANDAL (voice-over): Colossal is funding $175,000 of Yellowstoneús bioacoustic study.

Plus the company has hired a team of AI scientists to analyze the data the griz cams are collecting.

MATT JAMES: Weúre really hopeful then that they can collect tons and tons of data that our team can then begin to distill and train the AI to move on from just classifying wolf calls to classifying individual calls.

MATT STANDAL (voice-over): As weúve mentioned, all that data could include sounds and activities from people in the park.

The technology is so new that ethicists are still trying to understand the implications for human privacy in wild places like Yellowstone.

CHRISTOPHER PRESTON, University of Montana: This is all data that can be collected.

MATT STANDAL (voice-over): University of Montana philosophy professor Christopher Preston studies the ethics of human interactions with the natural world and how technology can shape those interactions.

CHRISTOPHER PRESTON: I mean, if you ask me, would I rather a wolf gets darted from a helicopter and have a radio collar put on it or a wolf gets listened to by a 24/7 recoding device?

Itús pretty clear to me that Iúd rather have the wolf be listened to by the recording device because thatús a noninvasive technique, much less likely to cause any sort of harm to the animal involved.

MATT STANDAL (voice-over): But Preston worries the cameras could inadvertently vacuum up human sound and images without people knowing theyúre being recorded.

CHRISTOPHER PRESTON: We do have a different sort of ethic for the human world to the one that we have for the wild world.

You go into landscapes like that to not be part of a system where people are looking at you, where people know what youúre doing and youúre certainly not getting away from it all if there is the potential for your movements to go into a database somewhere.

MATT STANDAL (voice-over): As Yellowstone experiments with this controversial new technology, biologist Dan Stahler is eager for more Griz Cams to be installed in the coming months.

He believes AI powered bioacoustics will help his team better protect these iconic animals as they learn more from every howl and that reverberates across this majestic landscape.

DAN STAHLER: Weúre going to keep this study going.

Thereúll be new emerging questions.

But the fundamental question will be why is Yellowstoneús wildlife community important to this landscape, important to Montana, and important to the world?

MATT STANDAL (voice-over): For PBS News weekend, Iúm Matthew Standal in Yellowstone National Park.