Zhao Chen Turn the special report to the Data Center business revenue growth page,
The words "CUDA≈App Store" were written in pencil in the margin of the page.
He raised his head and looked at Zhang Qingxue on the screen.
"Aunt Zhang, from a legal and compliance perspective, what knock-on effects will these optimistic expectations of Huang Renxun have?"
Zhang Qingxue lowered his head and pointed at several passages in the report that were marked red by Lin Shicheng.
"Lawyer Lin mentioned several key data in the special report--
There are two million CUDA developers, enterprise-level GPU acceleration customers have increased several times in a few years, and Data Center revenue has exceeded the game for the first time.
The legal significance of these data is that they are the first quantitative indicators disclosed by Huang Renxun in public.
Huang Renxun disclosed these data at the Financial Report conference call, which is tantamount to proactively providing the market with core parameters for valuation and pricing.
This has knock-on effects on several compliance levels.
First of all, Era Vanguard, as the largest institutional shareholder of Nvidia,
After knowing this non-publicly disclosed information, reverse operations cannot be carried out in the short term.
Now is not the time to sell. This is not only an investment judgment, but also a compliance judgment.
Secondly, after these data are digested by the market, the valuation center of Nvidia will most likely be repriced.
Once the stock price fluctuates significantly, the SEC may conduct a routine review of the Information Disclosure timing before and after the conference call.
We need to ensure that records of internal discussions and decision-making can withstand any subsequent scrutiny."
"What's your suggestion?"
"Lawyer Lin Make an information isolation memo to fully record the core conclusions and decision-making basis of today's discussion.
In the future, if the SEC inquires about Era Vanguard's trading behavior before and after the release of Nvidia Financial Report, this memorandum will be the first firewall.
In addition, Huang Renxun's optimistic expectations for the future--
'Computing power demand will grow millions of times over the next decade' --
A framework definition is needed at the legal level.
If these expectations fail to materialize as scheduled, Era Vanguard, as Major Shareholder that holds more than a certain proportion of shares,
May face potential challenges from other shareholders.
I suggest that in Investment Decision Committee's quarterly review report, Huang Renxun's expectations and Henglong Research Institute's independent Model be listed separately.
The former is labeled management guidance, the latter is labeled independent verification.
The difference range between the two is the core Risk Exposure that needs to be continuously tracked on Henglong on Nvidia.
Another point - Huang Renxun disclosed the number of two million CUDA developers for the first time, which legally constitutes public information.
But regarding CUDA ecosystem's competitive barriers, there are still some more detailed data that have not been made public.
For example, the industry distribution of developers, differences in penetration rates across regions, and cross-usage rates with major competitors.
This data may constitute material non-public information.
When Era Vanguard was communicating with the investor relations department of Nvidia,
Exposure to this type of information needs to be avoided, otherwise it will trigger insider trading compliance risks."
"Aunt Zhang, your opinion is very important.
Lawyer Lin, after the discussion, you and Aunt Zhang drafted this information isolation memorandum together.
The management guidance of Huang Renxun and the independent Model of Henglong Research Institute are presented separately, with the difference interval marked as core Risk Exposure.
This is a standard template for writing Investment Decision Committee quarterly review report."
Zhao Chen wrote down this key point on the paper, and then turned to Lin Shicheng who had been quietly reading the last chapter of the special report on the screen.
"Lawyer Lin, CUDA ecosystem's moat - you used a word in the report called 'living water', not 'moat'. Expand and talk about it."
Lin Shicheng Turn the special report to the last chapter.
He wrote this chapter all night long, and the source and verification path were marked next to each data.
He did not look at the slide on the computer screen, but directly raised the densely marked original analysis draft to the camera.
The red and blue marks on the page are densely arranged on the screen, like an engineering drawing that has been calibrated repeatedly.
"A moat is static - as long as the water doesn't flow away.
But CUDA is living water - two million developers write code, publish papers, start businesses, and get investment on it.
The innovative base for the entire AI industry is provided by CUDA.
Any competitor who wants to challenge Nvidia needs to copy not just the performance parameters of a chip.
It is a creative New Life ecosystem that has been circulating on its own for more than ten years.
There are several key figures for this ecosystem.
First, the proportion of the world's top AI papers citing CUDA as a computing platform has remained at a very high level for a long time.
This means that most of the most cutting-edge research results in academia are completed based on the Nvidia software stack.
Second, among the GPU acceleration instances provided by major cloud service providers, Nvidia has an absolute advantage.
Third, among the AI training tasks of large enterprises, CUDA-based Model also accounts for a very high proportion.
These three extremely high ratios constitute the core competitiveness of CUDA ecosystem.
Academia, cloud service providers and enterprise users - the three most critical computing power consumer groups,
All are deeply bound to the same software ecosystem.
Any competitor who wants to replace Nvidia needs to provide competitive alternatives in these three dimensions at the same time.
This is not impossible, but the time window and capital investment required are far beyond what most chip companies can afford."
Zhao Chen silently wrote down a few key data on the paper with Fountain Pen, raised his head and looked at Lin Shicheng in the video window.
"Lawyer Lin, the three dimensions you just mentioned are academia, cloud service providers, and enterprise users--
Can the independent Model of Henglong Research Institute be tracked separately?"
"Yes. But more research resources need to be invested.
Currently we are tracking comprehensive indicators--
The total number of CUDA developers, Data Center revenue growth rate, and the overall size of the AI training market.
If you want to split it into three independent dimensions for long-term tracking,
It is necessary to expand the research team of Research Institute and introduce more analysts with technical backgrounds."
"Then expand."
Zhao Chen said this decision very lightly, as if saying that today's tea is good,
But Chen Xi knew that this sentence meant that Henglong Research Institute's annual budget sheet would have to be made from scratch.
She quickly wrote a few lines in her notebook, then raised her eyes and looked at Zhao Chen.
"The three core judgments of Zhao Zong and Huang Renxun--
AI's iPhone moment, Data Center is the largest computing platform, CUDA moat--
We all made analysis from our respective fields.
But I think the biggest significance of this Financial Report is not to verify a specific judgment, but to verify the institutional framework of Henglong itself.
The position logic of Era Vanguard on Nvidia has gone through several major revisions from the beginning to today.
But every correction does not rely on someone making the decision at a critical moment, but on the cross-validation of Model, data and systems.
Every data mentioned by Huang Renxun last night is an independent verification of the effectiveness of our institutional framework."
Zhao Chen put down Fountain Pen, picked up Zisha Pot and poured a cup of tea into the cup.
"The position logic of Chen Xi, Era Vanguard on Nvidia has been verified one by one in today's discussion--
AI Computing Power's exponential growth in demand, CUDA ecosystem's moat, and Data Center's overtaking of the game.
Huang Renxun used public data to help us deduce these three logics.
In the current global allocation of Henglong Capital, the exposure of US Stocks is still very small.
If the position logic of Era Vanguard can withstand independent verification, Henglong can also be bought."
Chen Xi's fingers stopped on the keyboard.
She quickly understood the weight of Zhao Chen's words--
This is not arranging a specific transaction, but opening up the closed loop of Henglong System's internal investment logic.
In the past, although Era Vanguard and Henglong Capital belonged to the same actual controller,
However, the two are completely independent in terms of legal entities, capital channels and investment decisions.
For every position held by Era Vanguard on US Stocks, Henglong has never followed in the investment.
Now Zhao Chen offers to let Henglong also buy Nvidia,
It means that he has enough confidence in the position logic of Era Vanguard.
"How to control the buying scale and rhythm?" she asked.
"Follow Henglong Capital's own institutional framework.
Independent Model gave a buy signal, Investment Decision Committee passed the review, entered the market in batches, and the pace slowed down.
The position of Era Vanguard is not used as the anchor, but the independent verification of Henglong Research Institute is used as the anchor.
Era Vanguard is Era Vanguard, Henglong is Henglong.
Two independent legal entities, each making decisions according to their own institutional frameworks.
Nvidia took this ticket Era Vanguard for a long time, and the logic has been verified repeatedly.
If Henglong also wants to buy, it's not because Era Vanguard bought it.
It's because Henglong's own Model told the system that he can buy it."