In last Saturday's online exchange, readers again asked questions about Virtual. Moreover, with the continuous rise in Virtual's price and the ongoing success of Genesis Launches, Virtual has once again attracted the attention of many participants.
While focusing on the project's potential, it is indeed necessary to pay attention to the project's risks.
Therefore, in today's article, I want to share a more comprehensive view of this project, focusing not only on its potential but more importantly on my perspective on the project's risks.
Regarding its potential, I have already shared quite a bit in previous articles, mainly:
- The team is good, and the project continues to build and advance, even without interruption during the bear market, which is very rare.
- The project has certain profitability, and the project's vision is to create a "currency" for use in the ecosystem. In my view, this direction is correct, meaning the project's business model is good.
- Lastly, and most importantly, among AI + Crypto projects today, whether compared to similar competitors or other AI projects, its strength is the strongest, and this advantage tends to gradually increase.
From these perspectives, among the AI + Crypto projects I have seen so far, I believe it is the strongest.
However, at the same time, the potential risks facing the project are worth noting.
The biggest risk, in my view, is whether this approach of starting from ICO and combining AI with the crypto ecosystem through financial means is a long-term, stable, and sustainable approach for the AI + Crypto track?
To verify this, we still need to see more and stronger AI agents being incubated on the platform.
What constitutes a stronger AI agent?
Ideally, it would be an AI agent that breaks out of the ecosystem, used not only by crypto ecosystem users but also by traditional Web 2 ecosystem users. If this cannot be achieved, at least it should be an AI agent like Uniswap that must rely on blockchain technology to exist, which cannot be implemented by other technologies and solves a rigid demand in the ecosystem.
Currently, some well-known AI agents on the Virtual platform, including AIXBT in my view, do not quite meet this standard.
In fact, this risk has appeared in the past development of the crypto ecosystem, and I believe many readers have personally experienced it: the once-popular decentralized flagship project Filecoin.
Decentralized storage, in my view, has been and will continue to be a rigid need for human society, and the implementation of this application should theoretically rely on blockchain technology.
So when Filecoin appeared, it became an instant hit.
When the Filecoin project was first launched, its conditions were even better than Virtual's today - it immediately formed a dimensional attack on its competitors. It was almost flawless in technology, team, business model, and other aspects, and could be considered the cream of the crop.
But what happened?
After going online, although it was bustling for a while and indeed attracted some users for a period, as time went on, people discovered that after comprehensively considering efficiency, cost, reliability, and convenience, Filecoin's advantages were not obvious compared to its biggest competitor (various cloud services in the Web 2 ecosystem), and thus the project gradually cooled down, and subsequently, no large-scale user migration from traditional centralized cloud services to Filecoin was seen.
More notably, after Filecoin, another previously unknown decentralized storage project, Arweave, actually came from behind and surpassed Filecoin in terms of heat and demand.
When people compared Arweave and Filecoin, they found that Arweave's focus on permanent storage even better satisfied certain specific domain rigid needs to some extent.
Filecoin thus slowly became lukewarm as it is today.
Apart from using the specific example of Filecoin for comparison, we can also examine this risk from the opposite perspective:
If ICO is a suitable way of combining AI + Crypto, why has this method only shown effectiveness in launching AI agents, while there are few successful cases in other subdivisions of AI project financing?
In my previous articles, I had introduced another case I saw early last year: using ICO to launch the construction of small AI models.
This method is something I had been paying attention to earlier, but unfortunately, it was not very successful, and I have not seen more successful cases.
The now well-known DeepSeek is actually doing model construction work. They optimize open-source large models and develop their own model based on this. Besides their team's strength, another unavoidable reason for their success is that they earned a lot of money in quantitative trading early on, thus having sufficient financial resources to develop this model.
This shows that funding is an indispensable factor in language model development. If so, why has the ICO method been practiced in this field but never shown effectiveness?
I currently do not have answers to these questions, but I believe they are worth further verification and cautious observation.
If this risk truly exists, then Virtual's future would have a certain degree of uncertainty.
