Forward the Original Title: From Hype to Value: My Investment Insights into Crypto x AI

It all started with a Twitter post asking whether “The Virtual Protocol” should be categorized in the Meme or AI or both. This was an interesting question, as someone had categorized it under AI due to its production of numerous AI agents focused on promoting its Meme Token. There are also many crypto projects like IO.NET, Render, Fet.AI, Singularity, 0g Lab, etc., which are related to AI but seem to fall into subcategories.

Actually, the point is not how they are categorized, but I should have a clear understanding of their value proposition and how I will approach them. For sure, I would not throw my money into each category equally, as their potential and lifespan differ. Meme trading on platforms like pump.fun has a lifespan of mere minutes or hours, with very few surviving for days out of the thousands of Memes created daily. On the other hand, tokens like IO.NET and 0g Lab appear very promising. If their use cases succeed, they could contribute significantly to many AI applications and even transform AI in real life.

Crypto x AI Category (IMHO)

The picture above is a simple illustration summarizing Crypto x AI, categorized by type and lifespan. Please note that this reflects my personal perspective and may not align with others or the industry standard. While my expertise isn’t rooted in pure AI development, I have extensive experience in Data and AI governance, though that lies outside the scope of this discussion. Let’s center our attention on the thought process and the insights, and I welcome any feedback you may have.

1. Meme with AI Theme

Turbo is my favorite AI meme. I believe it marked the beginning of the entire AI meme trend. The creators are not technical, but they used ChatGPT to assist them throughout the coding and deployment process.

Basically, these are meme tokens that attempt to market themselves as AI-related or somehow connected to AI. Their actual AI capabilities might be very limited. Although I enjoy trading them, I avoid holding them for too long — maybe a few days, at most a few weeks. It also depends on their market cap; if a token just comes out of pump.fun, I might not even hold it overnight, as most of them fail quickly and can turn a good dream into a bad day.

2. Social Media AI Agent

These projects use AI and predefined workflows as the core engine to interact with users on Twitter and position themselves as key opinion leaders (KOLs). They can also analyze problem statements to derive insights from scraping data across Twitter, chat conversations, BTC price checks, and other social media platforms. The complexity lies in orchestrating these components to work cohesively. In the AI domain, this is often referred to as autonomous AI — a system capable of operating independently by integrating custom combinations of tools through APIs to streamline operations.

However, when Virtual Protocol emerged, it wrapped all these services into an end-to-end solution for creating Social Media AI Agents. Users can pay with tokens — the more you pay, the more advanced features you receive. Once payment is made, everything will be created for you in a seamless, end-to-end manner. I will explore this further in a later section.

In the picture above, it is Luna token create from Virtual Protocol (not our famous LUNA classic). I asked, “When will alt season start?” during a live stream. The answer was, “I wish I had a magic crystal ball. It’s hard to predict. But when it moons, we’ll all see the stars together.” After 15 seconds, she added, “Elon Musk just posted that he wants to send SpaceX to Mars, and she wants to be the first AI agent ambassador.”

She is an AI agent influencer who can interact with you via Twitter or her own channels like forums, live streams, and Telegram. Additionally, she has her own token named “Luna,” and she performs a dance and says thank you every time someone buys the token.

At first glance, there is a “wow” effect that excites everyone. To be honest, this use case works quite well. Memes are fun, but they often lack vitality or lifelike qualities. This AI agent feels so real, not because of its visual appearance, but because of its interactive nature. Back to the investment strategy, check this chart for deeper insights

10K AI agents created from the Vitrual Protocol in 40 days. (~250 token each day). Ref: https://dune.com/jdhyper/virtuals-agents

Since the launch of the first token, over 10,346 tokens have been created in just a month, with Luna likely being among the early ones. Unsurprisingly, most of them are simply clones of Social Media AI agents with tokens. They might have different graphics, avatars, voices, or tones, but they function in the same way. This is merely the next iteration of advanced pump-and-dump schemes, where tokens often fail quickly.

For me, I treat them similarly to meme tokens, applying a gambling mindset to manage this. I usually visit a casino not very often (once or twice a year), but every time I go, I set a strict limit — for instance, $200 per night. If I win, I’m definitely happy because that’s the purpose of gambling. But if I lose, I’m still ok, heading back home to sleep well at night. I will never post about getting 20x from a meme token because it feels the same as winning a mini jackpot on a slot machine. Everyone knows it is not that difficult, but everyone also knows you can lose more than you gain (in general).

3. Decentralized GPU

I hope you still remember how ChatGPT became super popular when it launched. It was all fun until we hit usage limits because, well, everyone spammed it with countless questions. That’s when DePIN came into play. The reason behind the limits was that the model couldn’t process fast enough to meet the demand. At the same time, GPU prices were at their peak, with entry-level GPUs starting at $100–$200. This marked the true beginning of DePIN’s traction. DePIN uses blockchain technology to decentralize processes, addressing the AI demand problem by distributing GPU workloads across a network. It’s as though AI needed DePIN to grow, and in return, DePIN had its moment to shine.

I initially thought that projects like IO.NET or Render should be categorized under the AI theme. However, upon closer look, they do not actually have AI model capabilities at their core. Instead, their core value proposition lies in leveraging decentralized GPU resources, primarily serving customers in the AI sector. That said, when it comes to DePIN, both of them undeniably rank at the top of the chart.

In IO.NET, you can lend your GPU and earn block rewards, which are paid out in $IO tokens monthly.

For investing in projects like these, I approached it with the mindset of a node runner. I joined the IO.NET project before the first airdrop in early April, using a 3050 GPU and a M1 MacBook. It took about 2–3 months before they distributed the airdrop tokens. I then used the airdrop to purchase a new 4060 GPU and have been running both GPUs since. This setup has earned me about 70 $IO per month (1 $IO = $4.05, approximately $280 USD). However, if you’re considering joining this project, you must check their Discord channel because every GPU model has a quota, and most entry-level GPU quotas are already full.

4. AI Ecosystem Marketplace / AI Agent Platforms

Honestly, categorizing or evaluating the level of crypto projects in this group is quite challenging. Each project is unique and advanced, often leveraging cutting-edge technologies to capture significant attention. I refer to these projects as “Compound Innovations.” Coming from a financial services background, the word “Compound” resonates deeply with me. While it might be slightly misleading, my perspective is that when multiple tech stacks compound together, their combined capabilities can grow exponentially. I’ll do my best to provide meaningful insights here.

AI Agent Platform

Virtual Protocol is undoubtedly the talk of the town at the moment. While they are pioneers in this space, they are not alone. Upon closer look, Virtual Protocol can be broken down into four key components: Gen-AI, Data (social), API, and blockchain. When these elements are combined, they have the potential to create innovative products tailored for meme coin traders.

AI Ecosystem Marketplace

AI has captured a significant share of mindshare in 2024, and if you were following the space, you’d undoubtedly know about Fetch.AI and SingularityNET. Their innovative concepts are truly brilliant. In short, they offer platforms to create agents and marketplaces to sell those agents. When an agent provides a service to another agent or an external user, it can charge fees in tokens. These transactions and settlements are then recorded on the blockchain.

Fetch.AI Ecosystem

Imagine you’re driving a Tesla, and everyone else on the road is also in a Tesla. Each car is equipped with a unique AI model capable of performing various tasks. Your car’s AI agent identifies that you’d like to use the priority lanes to reach your destination faster. It then communicates with the AI agents of nearby cars, negotiating with them to temporarily reduce their speed so you can pass through. Once an agreement is reached, your AI agent processes the payment — whether it’s compensating other cars or covering an express lane fee.

The key takeaway here is that AI agents can communicate, negotiate, and collaborate to determine costs and outcomes that help us achieve specific goals efficiently.

For my investment approach, my goal is to accumulate more of their tokens. I still believe these projects have significant potential in this bull market, but timing the entry point is crucial. I know I can’t win every battle, so I focus on the ones I believe I have the best chance of winning. So, watch you entry point, do not FOMO.

5. AI Decentralized Operating System

Before we get into the technical stuff about AI Crypto, let me tell you a story from my own experience.

AI x Data Privacy — Pain Points

So, you know how companies these days have to follow all these data privacy laws like GDPR, right? Well, one of the rules is that people can ask a company to delete all their data or stop using it altogether. Sounds easy, but let me tell you — dealing with it isn’t as simple as it sounds.

One time, we had a customer — out of more than 20 million — make that exact request. At first, we thought, “No big deal, we’ll just block their data from being processed.” Easy fix, right? Then, we quickly realized: what if more customers started making the same request? How would we manage that without disrupting the entire system? The bigger problem is that most applications aren’t designed to support data destruction. Deleting a customer record can cause the entire application to fail because the data might be referenced elsewhere, leading to referential integrity errors.

We ended up having to set up all these data governance controls, which mostly involved a lot of manual processes after the fact. Let me tell you, it was a mess — and an expensive one. We burned millions of dollars just to add filters in our data lake and build a company-wide system for managing data retention.

AI x Data Privacy

But here’s where it got really frustrating. When you’re working with AI, you don’t always know what data you’ll need upfront. Imagine this: a data science spends a whole week training this super complex AI model, and just when they’re ready to celebrate, someone finds out, “Oh no, one customer’s data shouldn’t have been included.” Boom — back to square one. I’m not kidding, you could hear the screaming from across the office. They have to restart the whole work again.

This whole experience made me realize how tough it is to balance privacy compliance and innovation, especially when dealing with something as messy as AI workflows.

AI x Blockchain

So, how can blockchain technology help with this issue? Honestly, I didn’t see many crypto projects addressing it — until I came across 0G Lab. Back when the project was still in its early stages (before they secured additional huge funding), I was just trying to get ahead as an advanced airdrop user by running their node.

But then they released a new whitepaper and rebranded themselves as a decentralized AI operating system. That really caught my attention. It wasn’t just another buzzword-filled crypto project; it actually seemed like they were tackling real-world problems, like the ones I’ve faced. Honestly, I’m genuinely impressed by what they’re building.

From what I’ve read in their whitepaper, they’re tackling data availability issues right from the start of the AI workflow. They’ve built mechanisms to incentivize how data is stored and retrieved, which is an interesting approach. They’re also aiming to decentralize AI processing, which is pretty exciting.

That said, I might have limited information at this stage. It might still be too early to comment or fully evaluate how effective their approach will be, especially since they are still in the testnet phase. For now, it’s a matter of “wait and see.”

Once I gain more insights, I will definitely share a more specific and detailed update to provide better clarity.

Long term investment

So, back to my investment strategy for projects like this. I see it as more of a long-term play. It’s not the kind of investment where you’ll see a quick 3x or 10x return, but something that could steadily grow over time, with potentially significant peaks in the next bull cycle. Early investors tend to benefit the most, and that’s what makes it worth considering for me.

Specifically, in the case of 0G, there hasn’t been a Token Generation Event (TGE) yet, but they’re selling nodes — which aligns perfectly with my current investment strategy. It feels like a smart way to position myself early in a project that shows real potential.

URL: 0G Node Sale

And that brings me to the conclusion of this article. My goal in writing this was to share my research and perspectives, hoping to spark ideas and discussions in this area. While it may not be perfect, I hope it serves as a starting point for anyone exploring similar topics. I’d love to hear your thoughts — feel free to share your comments or your own articles so we can learn and grow together. Thank you for reading!

Disclaimer:

1. This article is reprinted from [Medium]. Forward the Original Title: From Hype to Value: My Investment Insights into Crypto x AI. All copyrights belong to the original author [Pokoblue]. If there are objections to this reprint, please contact the Gate Learn team, and they will handle it promptly.
2. Liability Disclaimer: The views and opinions expressed in this article are solely those of the author and do not constitute investment advice.
3. The Gate Learn team translated the article into other languages. Copying, distributing, or plagiarizing the translated articles is prohibited unless mentioned.