AI helps decode whale language: Project CETI identifies 156 sperm whale codas
begusgasper · x · 2026-10-21
- Project CETI, a nonprofit founded in 2020, brings together 50 scientists from 8 institutions across AI, linguistics, cryptography, marine biology and robotics to decode sperm whale communication off Dominica.
- Analyzing nearly 9,000 recordings, researchers identified 156 distinct codas — click sequences of 3–40 clicks that serve as basic building blocks of the whales' language, with small variations potentially conveying different meanings.
- A Proceedings B study led by Berkeley linguist Gašper Beguš suggests sperm whales distinguish "vowels" via short vs. elongated clicks and rising vs. falling tones, showing parallels to human languages like Mandarin, Latin and Slovenian.
- The team also analyzed 61 hours of orca recordings (published spring 2026), identifying vowel- and consonant-like sounds, and deploys autonomous gliders to listen underwater.
More from AGI Musings
- Margaret Mitchell: click-driven AI misinformation is causing real harm — mmitchell_ai · 2026-10-21
- Mathematician Daniel Litt concedes his AI bet, but says AI won't solve math — ziv_ravid · 2026-10-21
- AI SaaS monetization is cooked: consumers small, enterprises pay $20/month — menhguin · 2026-10-21
- Tim Dettmers: Academia is about to have a renaissance — the best AI work of the next decade happens in university labs — StasBekman · 2026-10-21
- AI saves scientists 7 hours a week, but validation is the new bottleneck — AnneliesGamble · 2026-10-21
- UW's Amy X. Zhang brings CSCW theory to human-agent collaboration in Stanford talk — amyxzh · 2026-10-21