Using AI to ‘talk to animals’ might make us feel clever – but what, if anything, does it do for them? - theguardian.com

Marina: Welcome to Deep Voices — I'm so glad you're here. I'm Marina, a marine biologist who has spent the better part of my life completely enchanted by whales and dolphins, and this podcast is my way of sharing that obsession with you. Marina: Before we dive in — and we are absolutely diving in today — a warm thank you to our sponsors: Ada, who helps clients integrate A I into their workflows in smart, practical ways; Ago Consulting, pronounced "Ahh-Go," silicon development consultants working across everything from A I to SoC; and Zen Semiconductor, an umbrella company with a remarkable portfolio spanning the silicon industry. These are the people keeping Deep Voices alive, and I'm genuinely grateful. Marina: Now — today's episode. We've got an alphabet made of clicks. We've got dolphins in Florida with secrets that took fifty years to start unlocking. We've got a whale population so isolated it may be its own subspecies. And we're going to ask a question that I think deserves a real, honest answer: when we use A I to try to talk to animals, who is it actually for? Let's go. Marina: Okay. Thread one. Voices of the Deep. And I want to start in Sarasota Bay, Florida, because there is a study happening there that is — I'm not exaggerating — one of the longest continuous wildlife studies in the world. The Sarasota Dolphin Research Program has been following the same community of bottlenose dolphins since 1970. More than fifty years. They know individual animals. They know their mothers, their grandmothers, their calves. They know who hangs out with whom and who absolutely does not get along. Marina: And because of that extraordinary foundation — that depth of individual knowledge — researchers have been able to ask much more precise questions about how these dolphins actually communicate. A recent paper drawing on this long-term data has given us some genuinely new windows into dolphin vocal behavior. What they're finding is that dolphin communication is deeply contextual. The same animal can produce different signature whistles — those are the unique, personal whistles that function almost like names — in different social situations. Marina: Now, signature whistles. If you've heard about dolphin communication before, you've probably heard about these. Every bottlenose dolphin develops their own signature whistle in the first few months of life. They keep it for their entire lives. And other dolphins use it — they copy each other's whistles to basically call each other by name. That's not a metaphor. That's the leading scientific interpretation. Dolphins address each other individually using vocal labels. That's extraordinary. Marina: But what the Sarasota work is adding is nuance. It's not just: dolphin has whistle, dolphin uses whistle. It's: when does a dolphin use another dolphin's whistle? In what emotional or social context? What else is going on vocally at the same time? Because dolphins are producing a whole tapestry of sounds simultaneously — whistles, burst-pulse sounds, clicks for echolocation that may also carry social information. Teasing apart what's communication and what's navigation and what might be both at once — that's genuinely hard science. Marina: And the Sarasota team has the gift of knowing these animals as individuals. When you know that a particular female has a calf this year, and her mother is nearby, and there's a strange male approaching — you can start to ask whether her vocalizations change in that specific social configuration. That kind of fine-grained, longitudinal science is irreplaceable. It can't be rushed. It took fifty years to build. And what it's telling us is: dolphin communication is richer, more flexible, more socially sophisticated than we once imagined. Marina: There is something quietly moving to me about that. Fifty years of showing up. Fifty years of knowing the same animals, watching them grow old, watching their calves have calves. That kind of patient, devoted science — it's a love language in itself. And the dolphins of Sarasota Bay are the beneficiaries of it, and honestly, so are we. Marina: Alright. Thread two. Culture and Decoding. And this is where things get both thrilling and a little philosophically complicated. Let's talk about sperm whales. And let's talk about clicks. Marina: Sperm whales are the largest toothed predators on Earth. They dive to extraordinary depths — two kilometers, sometimes more — hunting giant squid in total darkness. And they do all of this while producing the loudest biological sounds ever measured. Their clicks are used for echolocation, yes, but sperm whales also produce something called codas — short, patterned sequences of clicks — that they use to communicate with each other. And the more scientists study these codas, the stranger and more wonderful the picture becomes. Marina: Here's what we know. Sperm whales live in matrilineal social units — groups of females and their young, led by the older females who carry the knowledge. And different groups, different clans, share different repertoires of codas. It's cultural. Groups that have been geographically separated for long periods have distinct coda dialects — patterns they learn from each other, pass down, maintain. Just like human cultures maintain distinct languages and songs. Marina: Now, for a long time, researchers thought of codas as relatively simple — maybe a few dozen types, used as identity markers. But Project CETI — the Cetacean Translation Initiative — has started to suggest that the combinatorial complexity of sperm whale codas is much, much greater than that. Project CETI is a massive, interdisciplinary effort using underwater hydrophones, A I, and machine learning to record, catalog, and ultimately try to decode sperm whale communication at a scale that's never been attempted before. Marina: And what their early analysis found is genuinely startling. When you apply the kind of computational tools that linguists use to analyze human language — looking at things like combinatorial structure, the way elements are combined and recombined to generate meaning — sperm whale codas show properties that are, well, language-like. There's something researchers call a "phonetic alphabet" emerging from the data. The whales appear to be varying tempo, ornamentation, rubato — that's a kind of timing variation — and rhythm in systematic ways that multiply the possible combinations far beyond what a simple inventory of coda types would suggest. Marina: The word "alphabet" gets used, and I want to be careful here, because we don't know what any of this means yet. We cannot say the whales are talking about their feelings or their squid or their grandmothers. What we can say is that the structural complexity of what they're producing is much greater than we thought, and that A I is the tool that allowed us to see it, because human ears and human pattern recognition simply cannot process this data at the scale and speed required. Marina: And then there's the question that an article in The Guardian raised recently — a question I think is really worth sitting with. When we build A I tools to decode animal communication, when we get excited about the possibility of "talking to animals" — who is that for? Is it for the animals? Or is it for us? Does it make their lives better? Does it help us protect them? Or does it primarily make us feel clever and connected? Marina: I don't think that's a cynical question. I think it's an essential one. Because there's a version of this science that is purely self-congratulatory, and there's a version that genuinely deepens our moral relationship to these beings. The researchers I most respect are asking it honestly. The hope — and it is a real hope — is that understanding what sperm whales are saying to each other will give us a window into their inner lives, their social needs, their responses to human disturbance from shipping and sonar and climate change. That it will make us better at protecting them. That understanding leads to caring, and caring leads to action. Marina: Project CETI is not just a translation project. It's ultimately a conservation argument. If we can demonstrate that these animals have something like language — something with genuine structural complexity and social function — then the moral case for protecting their acoustic environment, their migration routes, their family units, becomes almost impossible to dismiss. That's the bet. And I think it's a good one. But it has to be held honestly. Marina: And here's one more piece of this puzzle that I find quietly beautiful. The Arabian Sea. There is a population of humpback whales living in the Arabian Sea that almost never leaves. Most humpbacks are epic long-distance migrants — they travel from polar feeding grounds to tropical breeding grounds and back, thousands of miles every year. But the Arabian Sea humpbacks stay. They've been isolated from other humpback populations for potentially tens of thousands of years. And new genetic research from the Wildlife Conservation Society has now strengthened the case that they are a distinct subspecies — genetically different enough, behaviorally different enough, to warrant formal recognition as something apart. Marina: Why does that matter for communication and culture? Because isolation creates divergence. Just as human languages evolve separately when communities are separated, whale songs and dialects evolve separately too. The Arabian Sea humpbacks sing songs that are different from their Atlantic and Pacific cousins. Their culture has been shaped by their particular corner of the ocean, their particular social history, their particular challenges. They are their own world. And there are only a few hundred of them left. Recognizing them as a distinct subspecies isn't just taxonomic housekeeping — it's the first step toward giving them the specific, targeted protections they need. Science in service of survival. Marina: Okay. Thread three. A Cetacean and People Story. And for this one, I want to take you somewhere a little different. I want to talk about what happens when humans truly pay attention. Marina: The story of the Arabian Sea humpbacks is, in a way, a story about paying attention. For a long time, these whales were just — there. Fishermen in Oman and India and Pakistan knew them. They appeared in ancient coastal traditions. But the scientific world largely overlooked them, assumed they were just part of the broader humpback population passing through. Marina: It was careful, persistent scientific attention — tagging, genetic sampling, acoustic monitoring — that revealed the truth. These whales don't leave. They're homebodies in the most profound sense. The females give birth in the Arabian Sea. The calves learn their world there. The males sing their strange, isolated songs into those warm waters year after year. And the people who live along those coasts have shared that sea with them for millennia. Marina: There's something in that relationship — between coastal communities and the whales they've always known — that I find deeply moving. The local knowledge was there first. Fishermen could have told scientists: these whales are always here. They don't go. But science, when it finally caught up, gave that knowledge a different kind of weight — the weight that moves governments and international conservation bodies. Marina: And that's the conversation I think we need to get much better at having. Between traditional ecological knowledge and formal science. Between the people who've lived alongside these animals for generations and the researchers who arrive with hydrophones and genetic sequencers. The Arabian Sea humpback story is a case where those worlds are beginning to align, and the result is a stronger, better-evidenced case for protection. Marina: There are only a few hundred of these whales. A few hundred. In the entire Arabian Sea. That's a number that should stop you cold. These are not animals on the edge of extinction in some abstract, statistical sense. They are a specific community, with specific family bonds, specific songs, specific memories encoded in their culture and their genes. And they are hanging on. Marina: The threats they face are the threats that face so many cetaceans: fishing gear entanglement, vessel strikes, underwater noise from shipping, climate change disrupting their prey. The Arabian Sea is one of the busiest shipping corridors in the world. Every large vessel that passes through is a potential threat. And because this population is so small, so isolated, so genetically distinct — every individual loss matters enormously. Marina: Formal subspecies recognition — if it comes — would trigger protections under international law. It would make it harder for nations to dismiss their conservation as someone else's problem. It gives conservationists a legal lever. And it came from science. From the patient work of researchers who took a small, overlooked population seriously enough to study it deeply. Marina: I think about that a lot. The power of paying attention. The Sarasota researchers paid attention for fifty years and unlocked secrets about dolphin names. Project CETI is paying attention with every computational tool available to begin to decode what may be language. And biologists in the Arabian Sea paid attention to a few hundred humpbacks who never left home, and found something irreplaceable. Marina: Attention is not passive. It's a form of care. And in science, sustained attention is how we earn the right to say: we know something true about these lives. And knowing something true is the beginning of protecting it. Marina: So. Where does all of this leave us? We've traveled from a Florida bay where dolphins have been known by name — their own names, their whistle names — for fifty years, to the deep ocean where sperm whales may be speaking in an alphabet we're only beginning to see, to the warm waters of the Arabian Sea where a few hundred humpbacks carry a song unlike any other on Earth. Marina: What ties all of these stories together is something I believe deeply: that the ocean is full of voices. Real voices. Voices that carry meaning, that carry culture, that carry the weight of relationships and memories and identity. We are not the only beings on this planet who communicate with complexity and care. And the more honestly we reckon with that — not just as a scientific finding, but as a moral reality — the better we will become at sharing this world. Marina: A I is a remarkable tool for listening at scale. But it's still just a tool. The wonder has to come from us. The will to protect has to come from us. And I think, if we're paying attention, it will. Marina: Deep Voices, brought to you by Ada, Ago Consulting, and Zen Semiconductor. Thank you to all three for making this show possible. Marina: If today's episode moved you, or made you think, or made you want to learn more — please subscribe, share it with someone who loves the ocean, and leave a review if you're feeling generous. It genuinely makes a difference. And check the show notes for links to all the research and stories we talked about today. They're worth reading. Marina: I'm Marina. Keep listening for the deep voices. I'll see you next time.