The investment value of VIRTUAL

CN
7 hours ago

AI agents are a sector I am very optimistic about in this market cycle. Although the entire sector has fallen silent for various reasons, many projects within it are still continuously building and improving.

Virtual is one of them.

In this article, I will attempt to analyze the investment value of Virtual from a financial perspective.

As I mentioned in previous articles, information about project financial conditions in the current crypto sector is very scarce. Virtual is no exception. Moreover, the entire AI agent sector has only recently emerged, making this information even more limited.

The two pieces of information I found online that have some reference value are:

The first is the financial information publicly released by the Virtual team on their Twitter on December 31 last year (https://x.com/virtuals_io/status/1874111116403761316).

This information revealed that the annualized revenue of the Virtual project at that time was $300 million. The definition of this "revenue" is unclear; I estimate it refers to income. Since I cannot find further data, we will temporarily use "income" for estimation.

The second piece is a news article published on TradingView at the end of February this year (https://www.tradingview.com/news/todayq:1af82f429094b:0-virtual-protocol-revenue-drops-97-as-ai-agent-demand-declines/).

This news indicated that Virtual's daily income dropped from $1 million on January 2 this year to $35,000 on February 27.

To estimate Virtual's intrinsic value, we need to look at future free cash flow. To estimate future free cash flow, we can only rely on the existing free cash flow for prediction and linear extrapolation. However, the information at hand is insufficient to calculate the current free cash flow. Therefore, we can only reluctantly use "income" as a substitute for free cash flow in our estimation.

Last December marked the peak period for the AI agent sector, so the annualized revenue of $300 million indicated in the first piece of information is its peak value.

By the end of February this year, it was a low point for AI agents, with daily income of $35,000, which translates to an annualized revenue of $13 million.

I will still use 12 as the estimated value for "future PE." Therefore, Virtual's "intrinsic value" is approximately $160 million to $3.6 billion.

From this estimation process, we can see a significant gap between the optimistic and pessimistic valuations of its intrinsic value.

The main reason for this gap is the uncertainty surrounding the project's future development. This uncertainty can be further broken down into two aspects:

First, how will the future development of the "AI agent + Crypto" sector unfold?

Second, what will be Virtual's future positioning and development within this sector?

Regarding the first point, I have always been optimistic about its future development, but the uncertainty lies in whether the future application scenarios and business models of this sector will resemble what we see today.

In the Web 3 ecosystem, the current state of this sector is clearly not very good.

However, if we step outside the crypto ecosystem and look at the broader AI sector, we find that there are very few application projects in either Web 2 or Web 3 that have truly found a good business model. Even top projects like OpenAI currently have little to show in terms of commercialization, as their subscription revenue is insufficient to support their operations.

Among all AI companies, only Nvidia, as an infrastructure provider, has a solid business model.

So this may not just be a problem within the crypto ecosystem, but rather an issue for the entire AI sector.

Regarding the second point, among the many AI + Crypto projects currently, Virtual is one of the few projects I see that is profitable. The vast majority of other projects, regardless of how cool their described scenarios or advanced their displayed technologies are, have yet to show any significant profitability, let alone boldly disclose their revenue status.

Therefore, in the current AI + Crypto sector, there is a great deal of uncertainty regarding Virtual's future revenue situation, but looking across the entire sector, I don't see many projects with better certainty than it.

We have roughly estimated Virtual's intrinsic value, and now we need to look at its "price." Here, we can only examine its token.

Regarding the valuation of the project token, I have expressed my views in several previous articles:

The value of a pure governance token is extremely limited; only tokens that become truly valuable in terms of rights or become essential "goods" for the project have value.

In this regard, Virtual has clearly stated its future intention to develop the token into a currency used daily within the "nation" of AI agents, and it has been implementing similar practices in its ongoing operations.

Whether this goal can be achieved will only be proven by time, but the effort to move in the direction of "utility" is the right approach. Therefore, in this respect, it is much stronger than many long-established projects.

Finally, let's assess whether the current token price is overvalued.

As of the time of writing, Virtual's fully diluted market capitalization is $580 million. If we calculate using the lower limit of the "intrinsic value" of $160 million, it is indeed overvalued. Therefore, I will not buy more. However, for such a sector and such a project, I am willing to give it a bit more patience, so I will retain the tokens I hold.

免责声明:本文章仅代表作者个人观点,不代表本平台的立场和观点。本文章仅供信息分享,不构成对任何人的任何投资建议。用户与作者之间的任何争议,与本平台无关。如网页中刊载的文章或图片涉及侵权,请提供相关的权利证明和身份证明发送邮件到support@aicoin.com,本平台相关工作人员将会进行核查。

Gate:注册解锁$6666
Ad
Share To
APP

X

Telegram

Facebook

Reddit

CopyLink