/
Navigation
Chronicles
Browse all articles
Explore
Semantic exploration
Research
Entity momentum
Nexus
Correlations & relationships
Story Arc
Topic evolution
Drift Map
Semantic trajectory animation
Posts
Analysis & commentary
Pulse API
Tech news intelligence API
Browse
Entities
Companies, people, products, technologies
Domains
Browse by publication source
Handles
Browse by social media handle
Detection
Concept Search
Semantic similarity search
High Impact Stories
Top coverage by position
Sentiment Analysis
Positive/negative coverage
Anomaly Detection
Unusual coverage patterns
Analysis
Rivalry Report
Compare two entities head-to-head
Semantic Pivots
Narrative discontinuities
Crisis Response
Event recovery patterns
Connected
Search: /
Command: ⌘K
Embeddings: large
TEXXR

Chronicles

The story behind the story

days · browse · Enter similar · o open

How researchers are using self-supervised learning, a branch of AI that has proven effective for handling human language, to better understand animal sounds

Wall Street Journal Christopher Mims

Context & Ripple Effects

The Wall Street Journal piece is the earliest marker in a research arc that ran through 2022–2024: applying self-supervised learning — the technique behind modern language models — to animal vocalizations. Five months later, the New York Times reported scientists decoding communication in fruit bats, crows, whales, and naked mole rats, with some eventually aiming to converse with marine animals.

The significance goes beyond bioacoustics. Quanta Magazine reported computational neuroscientists finding self-supervised models correspond more closely to brain function than supervised ones, and the same learning-from-unlabeled-signals logic later showed up in Meta's AI lab decoding EEG readings and in models trained on infant headcam footage to study language acquisition. Animal sound is one instance of a broader move: AI learning structure from raw, unlabeled signals.

First-order effects

  • Researchers studying animal sounds gain a method that works without labeled datasets — the bottleneck that previously limited bioacoustics to species where humans had already annotated calls.

Second-order effects

  • The same self-supervised toolkit spreads across adjacent signal domains: the NYT's animal-communication work, EEG decoding at Meta's AI lab, and infant language-acquisition studies all reuse the approach, turning bioacoustics into a proving ground for cognition research more broadly.

Third-order effects

  • If self-supervised models keep tracking brain function better than supervised ones — as computational neuroscientists told Quanta — the field's center of gravity shifts toward learning from raw sensory streams, with animal communication, neural signals, and child development studied under one methodological umbrella.

The trend: AI research is extending self-supervised learning from human language to every raw signal domain — animal vocalizations, brain activity, infant speech — as a unified method for decoding natural communication.

Discussion

  • @julespolonetsky Jules Polonetsky on x
    It's clear we will soon understand animal speech/intelligence and communication far better. I imagine this will lead to much more concern for animal welfare, vegetarianism and environment? Imagine the transcripts published in media. 1/2 https://twitter.com/...
  • @julespolonetsky Jules Polonetsky on x
    Or more likely, the world will only pay attention to certain conflicts, but not others, as we do with human strife.
  • @etzioni Oren Etzioni on x
    Fascinating: Alexa for Animals https://www.wsj.com/... by @mims
  • @drescotet Miguel Angel Escotet on x
    Alexa for Animals: AI Is Teaching Us How Creatures Communicate. New kinds of artificial intelligence are enabling scientists to better understand the sounds of the animal world, from whale songs to mouse squeaks • #Psychology #AI #animalpsychology • https://www.wsj.com/... • http…
  • @2morrowknight Sean Gardner on x
    Researchers are using #AI to parse the “speech” of animals, enabling scientists to create systems that, for example, detect and monitor whale songs to alert nearby ships so they can avoid collisions. https://www.wsj.com/... #ML #DeepLearning #NLP #IoT #animals @WSJ #DYK
  • @mims Christopher Mims on x
    Thanks to everyone who weighed in, I couldn't in the end resist how iconic John Krause was able to make this illustration, & how well it fit the subject, so we went with the RCA reference https://www.wsj.com/... https://twitter.com/...
  • @mims Christopher Mims on x
    With “Alexa for Animals”, which uses AI to parse the “speech” of animals, scientists are asking a profound new question: Is the best way to probe one alien intelligence to use another? https://www.wsj.com/...