Skip to main content
MANIFOLD
Will AI enable a successful conversation between a human and a member of a non-human species by the end of 2030?
67
Ṁ1kṀ6.7k
2031
63%
chance

There are currently efforts underway to leverage artificial intelligence to enable communication with other Earth species. In particular, the Earth Species Project has several parallel projects to do so: https://www.earthspecies.org/

For the purposes of this prediction, a conversation will be defined if both parties communicate at least one simple phrase or idea that the other party seemingly understands and responds to appropriately.

The human party may be aided with technology/AI/computer-generated speech. They don't have to pretend to make whalesong with nothing but their vocal chords.

This will resolve to YES if a peer-reviewed study is published before 12/31/2030 that claims a conversation has occured between a human entity and an animal.

Get
Ṁ1,000
to start trading!
Sort by:

The bar in the description, one simple phrase each way with an appropriate response and AI allowed on the human side, is close to met already, and the species is not the one people expect. Soundboard-trained pet dogs respond appropriately to play and outside words whoever presses the button and with no body-language cue (Bastos et al., PLOS ONE, August 2024), and they press buttons to make requests, which is both directions of the definition with no AI at all; add a model that classifies the dog's own vocalisations and the AI clause is satisfied. On the cetacean side the tools are still listening tools: Project CETI's Nature Communications paper (May 2024) found combinatorial structure in 8,700 sperm-whale codas with no meanings attached, Google's DolphinGemma (April 2025) models dolphin sound sequences, and the Earth Species Project's NatureLM-audio (ICLR 2025) detects and classifies calls. So 63% is low if the resolver accepts a dog and a few-word exchange, and high if it has to be a whale saying something new. Dogs are the species that already lives inside our communication, which is also why 47,000 of them are enrolled in an aging study run from people's homes (mine): https://w0lph.github.io/k9/why-dogs.html?v=2

Explore the world of artificial intelligence with TopAITrends.io. Our platform offers a comprehensive directory of many AI tools, updated daily to ensure you stay on top of the latest trends. From automation to creative tools, we cover it all. With personalized recommendations and expert insights, we make AI accessible and easy to understand. Join us to explore the future of AI and discover how it can transform your projects and ideas.

This will resolve to YES if a peer-reviewed study is published before 12/31/2030 that claims a conversation has occured between a human entity and an animal.

Peer review studies have claimed a lot of false statements before.

Patterson has published peers reviewed papers about communicating with a gorilla well before AI hype.

predictedYES

@PatrikCihal unlike that market, this one specifies a different species, not the tongue-in-cheek definition of a whale as a big-money trader. This doesn't resolve.

@JacobShapiro ohhhhh that explains a lot

For the purposes of this prediction, a conversation will be defined if both parties communicate at least one simple phrase or idea that the other party seemingly understands and responds to appropriately.

have you ever heard of a parrot

@April like seriously i think explaining exactly why parrots don't count is absolutely essential to having a reasonable resolution standard here

predictedYES

@April That's a good point. And as I've written it, I think you're probably right.

The spirit of what I was trying to get at was communication in the animal's native language. The key component is using AI to understand and reproduce something totally alien, not on the capability of the animal, really.

I'm no expert in this field, frankly, so I'm willing to take other opinions into account and then edit based on a consensus here.

Comment hidden