The global AI market already exceeds $2.5 trillion in estimated spending for 2026. But what's interesting isn't just the size of the number — it's what companies are actually looking for today: not specialists in a single technology, but people capable of combining AI, Cloud, Blockchain, security and automation. That integration vision isn't new. There were projects that tried to combine AI and Blockchain long before it became trendy, and it's worth looking at what happened to them, because what they built — and what they failed to achieve — is still relevant today.
Although blockchain is a much smaller sector than AI, the trend we're seeing is that companies aren't only looking for AI specialists. They're looking for people capable of integrating:
- AI
- Cloud
- Security
- Data
- Blockchain
- Business automation
Sure, AI moves way more money than blockchain by a huge margin. But blockchain remains a strategic technology in areas where trust, auditing, identity, traceability and digital assets are needed. That's why many companies end up needing both instead of choosing one or the other.
DeepBrain Chain is a project I found interesting precisely because it tried to combine two things long before it became fashionable: AI + Blockchain. This idea was born in 2017, with the proposal of creating a decentralized computing power network, mainly GPUs, so that AI companies and developers could rent computational resources cheaper than those from traditional providers.
If you think about it, the concept is pretty similar to what several decentralized computing networks for AI are trying to do today.
So, is it a success?
Its market had a lot of hype during the 2017-2018 crypto boom. Its all-time high was close to $0.66 USD per token — today it sits around $0.0003 USD, more than 99% below that peak.
This doesn't automatically mean the project is a fraud, but it does indicate it never reached the adoption many investors expected, for several reasons:
It arrived too early
The AI market of 2017 was much smaller than today
Decentralized GPU infrastructure was hard to build
It ended up with a pretty damaged reputation within the crypto ecosystem. But the curious thing is that its idea doesn't seem so crazy in 2026.
What happened to Ocean Protocol? Does it still collaborate with car manufacturers?
Ocean Protocol was a fascinating case and yes, it still exists, but it's no longer in the position it held between 2020 and 2024. And the famous collaborations with car manufacturers don't seem to be the center of the story today.
Ocean wanted to create a decentralized data economy — a very brilliant idea, since companies own data, don't want to sell it completely, but do want to monetize it. Ocean allowed data access while maintaining a certain level of control and privacy.
That's how it caught the attention of industrial and automotive sectors. It participated in initiatives like Catena-X alongside companies from the European automotive ecosystem and organizations like Bosch and Fetch.ai.
When you think about supply chain, traceability, predictive maintenance or connected vehicles, that's where Ocean made sense.
With the AI boom, Ocean got very close to these projects, giving birth to what became known as the Artificial Superintelligence Alliance (ASI). The idea was to unite AI + Blockchain ecosystems under a common vision, and at that moment a lot of people thought:
"They're building the decentralized infrastructure for the AI economy"
And here comes the twist.
In 2025, Ocean left the ASI alliance, going back to operating independently. It stated it wanted to regain control of its tokenomics and strategic direction. There were also cross-accusations and governance conflicts among the participants.
The interesting thing is that Ocean hasn't disappeared. The project's own documentation still describes Ocean as an independent token after the ASI exit, and the team continues developing a vision around decentralized data and AI.
In crypto communities there are still people who see it as a historical data infrastructure project for AI that's trying to reinvent itself after the break with ASI.
Its original idea remains extremely powerful — and not because it's a cryptocurrency, but because it raises a question that companies still haven't fully resolved:
"How do you share data between organizations without losing control over it?"
From the technical side, there are concrete answers: Hyperledger, APIs, messaging systems, Smart Contracts coded to unlock information only with the private key of whoever is meant to receive it. The tools exist. But The Matrix Revolutions left us something in one of the last dialogues between The Oracle and The Architect: technical problems tend to be hard, but human problems tend to be even harder. And in the business world, sharing data between organizations isn't just a technical problem. It's a problem of trust, incentives, governance and interests that don't always align. That's where projects like Ocean Protocol tried to build something that goes beyond the code.
I wouldn't say Ocean Protocol revolutionized the automotive industry, but it was one of the projects that helped popularize the idea of decentralized data markets and data sharing between companies — something with very important applications in the automotive sector.
And finally, let's talk about the Cortex project
Because it was one of those projects that, when it appeared, seemed like science fiction. Its proposal was:
"What if Smart Contracts could execute artificial intelligence models directly inside the blockchain?"
In 2018 that was an idea very ahead of its time. Cortex wanted to upload AI models to the network, execute inferences inside the blockchain, enable smart contracts with AI capabilities, and create AI DApps — better understood as decentralized applications with integrated AI.
They claimed to be the first blockchain capable of supporting on-chain AI, and in terms of marketing and architecture, they had a basis for that claim. Even today Cortex still describes itself as the first — and according to them, the only — public blockchain capable of executing AI inference directly on-chain.
Now imagine a smart contract that asks:
Does that transaction look fraudulent?
Was the payment received?
Did the goods arrive before the deadline?
But those questions are answered by the smart contract itself without needing AI — it just has to be fully configured, or in other words, the rules have to be defined.
And that's one of the reasons why thousands of AI+Blockchain projects ended up having less impact than they promised.
Everything can be programmed with traditional logic. AI starts to make sense when the answer can't easily be expressed through rules — for example, when probabilistic inference is involved.
Cortex still exists as an active project and its documentation still claims the title of first public blockchain capable of executing AI inference on-chain. But it never reached mass adoption, partly because it arrived too early and partly because the problem it was trying to solve turns out to be unnecessary in most real-world cases. Smart contracts don't need a classification model to know if a payment was received. Traditional logic handles that perfectly. Cortex found the right answer to a question very few companies were asking.
Many times, especially in blockchain and AI, a tendency appears to think that if a technology is good, it should do everything.
In a symphony orchestra, nobody says: the violin is so good it should replace the piano, or percussion is so important that all music should be percussion.
Every instrument has strengths and limitations — and it's the same with technology. Blockchain is excellent for trust, traceability, auditing and coordination between actors. Traditional databases are excellent for speed and massive storage. AI is excellent for classification, prediction and analysis. APIs are excellent for integration. Cloud is excellent for scalability. Messaging systems are excellent for communication between services.
Top comments (0)