Bittensor’s TAO Takes a Jaunt After Huang’s Rave: Is Crypto the New Black?

by

Well, blow me down! Bittensor’s TAO has decided to don its Sunday best, soaring like a particularly enthusiastic kite on Thursday and continuing its merry jig into early European trading on Friday. The catalyst for this sudden enthusiasm? None other than Nvidia’s own Captain of Industry, Jensen Huang, who waxed lyrical about the project on the All-In podcast. As a result, the token leapt from a rather pedestrian $243.5 to a dizzying $310.6 before taking a breather at $298.1 when we last checked in.

Now, let’s not get too carried away, dear reader. Huang’s endorsement wasn’t a direct pat on the back for Bittensor itself, but rather a nod to a technical milestone that sparked much chatter over the vast ocean of open AI infrastructure, which is about as hot as a freshly baked scone at tea time.

The whole kerfuffle began when Chamath Palihapitiya, with all the fanfare of a ringmaster at the circus, directed Huang’s attention to what he dubbed a “pretty crazy technical accomplishment” within “this crypto project called Bittensor.” He was referring to some sort of recent training run on Subnet 3, where participants utilized distributed excess compute to train a Llama model-yes, a Llama, because why not-“totally distributed” while still managing to keep everything shipshape in a stateful manner. One can only imagine the chaos if it were otherwise!

Nvidia CEO Responds To Bittensor’s Accomplishment

Huang’s response was as succinct as a good cocktail recipe: “Our modern version of Folding@home.” And there you have it, folks! That little quip transformed Bittensor’s latest achievement into something that even your Aunt Mabel, who still thinks Bitcoin is a type of cheese, could understand. Folding@home was one of the earlier examples of decentralized volunteer computing, and Huang’s comparison suggested he viewed Bittensor’s doings less as a crypto farce and more as a genuine expression of distributed coordination-a term so highfalutin it could make a cat look sophisticated.

In the midst of this price action, traders took Huang’s remarks as a sign from the heavens, interpreting it as external validation from a heavyweight in the AI hardware arena. Who wouldn’t want a thumbs up from someone who holds the keys to the kingdom?

But Huang didn’t stop there; oh no, he decided to expand the conversation into the labyrinthine structure of the AI market. “I believe we fundamentally need models as first-class products, proprietary products, as well as models as open source,” he proclaimed with the flair of a Shakespearean actor. “These two things are not A or B, it’s A and B. There’s no question about it.” If only life were so simple, eh?

it’s 72 billion parameters, not four. Trained permissionlessly across 70+ contributors…

– templar (@tplr_ai) March 19, 2026

However, Huang drew a hard line when it came to industry-specific deployment, insisting that domain expertise “has to be captured in a way that they can control,” and that “it can only come from open models.” A classic case of wanting to have one’s cake and eat it too, if there ever was one.

This distinction goes straight to the heart of Bittensor’s wild reaction. While Huang didn’t officially crown any token as the king of open AI, he did endorse the harmonious coexistence of proprietary and open model ecosystems, while acknowledging that specialized industries would require more controllable, open foundations. It’s a curious world we live in, where even the virtual currencies can feel the thrill of approval.

At the time of writing, TAO was trading at a rather respectable $297.0, proving once again that in the land of crypto, the only constant is change-much like the weather in England.

Read More

2026-03-20 15:58