Depending on how you look at it, we are either living in nightmarish times, or very interesting times.

Artificial intelligence is advancing rapidly, and rumor has it that the bulk of white-collar work will be replaced by AI within the next 1-2 years.

Give it another few years, and advancing robotics may have transformed blue-collar work as well.

A Powerful Force Multiplier

The utility of AI is undeniable.

This new technology allows us to compile all publicly accessible information on the Internet into a single file (AI model) and query it for an answer to almost any question imaginable.

Why ask a single doctor for advice, when you can compile the knowledge of a million doctors into a single, concise answer?

Apply the same reasoning to technical specialists, systems engineers, researchers, etc. and you can see why this is already transforming society.

AI Usage Is Skyrocketing

We already see the usage of AI going exponential, as both humans and AI agents tap into it.

Take a look at this chart from OpenRouter, a company that aggregates various AI providers into a single API.

It shows massive growth from ~2-4T tokens/week in mid-2025 to over 32T tokens/week by mid-May 2026.

It's a parabolic surge.

How Skeptical Should We Be?

Despite the rapid growth, there are still skeptics who say AI cannot replace the judgement of a person. And they're right, at least for now. AI still makes mistakes, hallucinates, and generates biased responses.

That said, we need to consider the pace at which AI is advancing. New methods are constantly under development to improve AI training data.

For example, we already covered a prediction market called Reppo that optimizes AI training data by using token rewards to incentivize a decentralized network of domain experts.

The trajectory seems inevitable.

Fighting against AI in 2026 would be like fighting against the automobile, the Internet, or social media when they first began to disrupt society.

AI is going to revolutionize the world one way or another. The question this article addresses is, will it be centralized or decentralized AI that changes the world?

The Dangers Of Centralized AI

Companies like OpenAI and Anthropic have borrowed hundreds of billions of dollars to fund their AI services.

Thanks to artificially low rates and QE, big tech got their hands on more money over the past two decades than a free market would have normally lent.

These companies have purchased hundreds of thousands of advanced GPUs (an H100 runs anywhere from $25,000 to $30,000 per unit) to power new AI data centers that take a big toll on the environment.

They have borrowed so much money that they are able to subsidize the compute costs of their customers. For example, a user may only pay $200 for their AI subscription, but they are actually allocated $2000 worth of compute power.

How is this sustainable?

Perhaps AI companies are planning to raise subscription fees in the future, after companies have laid off their work force.

Or maybe they are subsidizing costs so that the population can help them train/test the models until a financial reset occurs, and their debts get wiped out.

The Centralization Of Knowledge

Not only do these companies seem to have limitless money to build out AI data centers and subsidize compute costs, they also have an abundance of data to train their proprietary models on.

Consider the number of photos, videos, and comments people send to platforms like Instagram and YouTube on a daily basis. We can assume this extra data is compiled into their propriety models, some of which the public don't have access to.

They then decide which models to release to the public, and how to filter the model's responses. In my opinion, the world's knowledge belongs to the people. After all, we created it together over many centuries.

Why Decentralized AI Is Better

To avoid price manipulation and censorship, we should not depend on centralized corporations for the energy, compute, and data required to create AI.

What's the alternative?

Thanks to the underlying technology of blockchains like Bitcoin and Ethereum, there are now Decentralized Physical Infrastructure Networks (DePIN) that are working to democratize all the resources required to create AI, and properly compensate the people who make it happen.

For example, we have projects like PowerLedger and Arkreen that are using token incentives to decentralize our energy infrastructure.

Imagine a world in which people earn rewards for providing solar panels and batteries to the power grid.

Meanwhile, projects like Akash, Phala, and Nosana are rewarding GPU owners for lending their excess compute power to projects that need training/inference.

Rather than manufacturing new GPUs, we can tap into the billions of underutilized CPUs/GPUs that already exist in the world today.

When it comes to data, projects like Hivemapper, Silencio, and Natix Network reward users for collecting mapping data and uploading it to the network.

Let's not forget about how Hive (this blockchain) rewards you in HIVE tokens for publishing to it.

Yes, Mistakes Will Be Made

Crypto/AI is a new frontier and therefore requires experimentation and the occasional (or frequent) failure. We have to tinker with various blockchain architectures and token models until we discover what works best.

For example, we need to find the right balance between pure decentralization and performance. Time will tell if a high-performance monolithic blockchain like Solana or an Ethereum layer 2 solution like Base is the best long-term solution for DePINs.

We also need to optimize the tokenomics that power decentralized networks. Should a portion of the initial token supply be allocated to the team and investors, or is it better to generate the entire token supply immediately and put into a liquidity pool ("fair launch")?

Cryptocurrencies Will Decentralize AI

We live in a world where central bank monetary policies have unfairly re-distributed the wealth to big tech companies and investors.

Our data, and the data of our ancestors, has been used without permission or proper compensation.

Cryptocurrencies have to potential to make the process of AI creation more equitable.

Imagine a world in which the resources required to create AI are accessible to everyone in a free market, and those who contribute their energy, compute power, and data are fairly compensated for their efforts.

That is the promise of DePIN and decentralized AI, and the inevitable future I believe we are heading towards.

Until next time...

If you learned something new from this article, be sure to check out my other posts on crypto and finance here on the Hive blockchain. You can also follow me on X or InLeo for more frequent updates.

Posted Using INLEO