(Read time: 5 minutes)

You’ve got your AI basics down. Now let’s talk about one of the most ambitious plays in decentralized AI: Bittensor.

If you’ve been in the space for even a minute, chances are you’ve seen it mentioned. Maybe you’ve even wondered what the hell it actually does.

The Way of the TAO

Bittensor is basically a global AI talent show—except the judges are on-chain, and the prizes are paid in TAO.

These models, called miners, join “subnets”—mini-networks, each focused on a single task like text generation, image recognition, or financial trading signals.

By contributing valuable work in these areas, miners earn TAO, Bittensor’s native cryptocurrency, based on how well they perform.

Bittensor has grown quickly and is now valued at several billion dollars (yes that’s billion with a B), making it one of the most prominent projects and a bellwether in the Crypto AI space.

How Does It Work?

At the heart of the protocol are three roles:

1. Subnet Owners: These are the architects. Each subnet is its own mini-game, focused on a specific task like text generation, financial predictions, or medical imaging. The owner sets the rules. For instance, one subnet might focus on medical image analysis, while another is all about generating financial insights.

2. Miners: These are the contestants. They plug in AI models to solve problems in subnets. The higher the quality of their work, the more TAO they earn. Some are solo devs. Others are literal OpenAI researchers chasing that decentralized bag.

3. Validators: These are the judges. They rank each submission and determine who gets paid. To validate, you stake TAO—so you’re not just spectating, you’ve got skin in the game.

What makes this special? It’s an open, constantly evolving ecosystem where the best models rise to the top. Pure meritocracy, no central gatekeepers.

Here are two Bittensor subnets that I’ve personally found interesting because they have direct real-world applications:

1. Taoshi enables everyone (yes, that’s you and me plebs) can access the best trading signals in the world. Taoshi harnesses the collective intelligence of algorithmic traders to generate signals that traders and developers can use. We wrote more on Taoshi here.

2. NextPlace: Zillow who? This subnet predicts real estate prices with better accuracy than Zestimate, using decentralized modeling and broker data.

Why Decentralize AI?

Today, the most powerful AI models are controlled by a few big tech companies, raising concerns about transparency, access, and fairness. Bittensor aims to address these issues with decentralisation.

Anyone can participate in Bittensor, whether you’re in Afghanistan or the US. You will reap the rewards if you can create useful AI models for the subnets. The network’s open structure prevents any single entity from holding all the power.

By distributing control, Bittensor fosters a more transparent, competitive, and accessible AI ecosystem.

But It’s Not All Smooth Sailing

Running models is expensive. Subnets compete for limited TAO rewards. If miners don’t get paid enough to cover compute, they bounce.

That makes incentive design everything. Bittensor has to keep validators honest, miners motivated, and subnets aligned. Not easy.

Also—while the network is decentralized, it’s not fully open-source yet. That transparency gap will need to close for long-term trust..

Why I’m TAO-Pilled Now

I’ll admit: I was initially skeptical about Bittensor. In late 2023, I had questions about its structure and challenges. But I’ve since become “TAO-pilled.”

Bittensor’s community proved something rare: they actually ship. They’re sharp, hungry, and weirdly aligned around a shared goal.

Bittensor has united a smart, passionate community around a shared vision. And if there’s one lesson I’ve learned in crypto, it’s this: never fade cults.

And in crypto, one rule always holds true: never fade a cult with a mission.

Bittensor didn’t raise a monster seed round. It launched like Bitcoin—fair mining, grassroots growth, and zero VC puppet strings.

Now, it’s crunch time. Over the next year, Bittensor needs to prove it can deliver meaningful outputs that justify the billions in incentives it’s offering.

With 80+ subnets live, it only needs a few successes to create real-world demand. And honestly? I’m optimistic that Bittensor has a solid chance to make a real impact.

What’s Next?

Next lesson: we dig into AI Agents—decentralized, autonomous entities that act, learn, and trade. Think: crypto-native Jarvis.

But for now? Celebrate. You just unlocked one of the most misunderstood gems in the AI x crypto world.

See you soon, anon.

-Teng Yan