Chapter 349: More Expectations for Artificial Intelligence (4/4)

There are many such examples, and they can further decompose the functions of their own application software and provide more independent services to the outside world.

Thinking of this, Ye Zishu said: "Actually, many subsidiaries of Phoenix Technology have unique technologies that can make independent cloud service functions.

For example, the graphics processing software of Jiufeng Software Company can deconstruct many functions in it, such as their picture text recognition function, which is a very good function.

After the cloud service function is made, other development companies can purchase this service separately and use it in their own products to form a whole to provide external services.

There are many other similar ones. For example, the function of automatically equipping text according to the sound in your video processing software can also be made into a separate cloud service.

Since there are too many, I will not give examples here. You need to conduct careful internal research. In addition to deconstructing existing software technology functions, you can also independently develop more independent cloud services. "

Ye Zishu's words opened up new ideas for Lei Jun and the others. It turned out that cloud services can still be used in this way, and there is no need to provide complete software at all. Instead, many software functions are disassembled and sold separately.

In this way, not only can additional sources of business income be obtained, but the value of the services provided individually is even higher than the value of the overall services provided.

More importantly, when more and more enterprises use their cloud functions for software and Internet application development, in order to improve the overall performance, they have to deploy services on the same cloud platform.

Even worse, these functional services only serve the applications deployed on this platform, and the use of other cloud platforms will be restricted.

This will produce a cluster effect, and everyone will unconsciously gather in a cloud computing platform, call each other's services, and form a cross-network structure on this platform.

Once this situation is formed, the cost to replace the cloud computing platform will be very high. After comprehensive consideration, many enterprises will not change the platform casually, even if other platforms are cheaper, they will not migrate.

Although Yeshu wants to dominate the global software market, he knows that it is impossible to rely on a single company. It can only be said to occupy the majority of the software market.

Therefore, he will naturally not object to opening up more functional services to the outside world, because this is the general trend, and he cannot stop it.

Unless all the brooms are self-preserved, the cloud computing platform is quietly used internally and not used by the outside world, but other people are not fools, knowing that there is such a technology,

Naturally, there will be companies doing research.

However, it is only for internal use in the initial stage. On the one hand, the benefits of providing external services are not very large, and the current technical architecture has not been widely recognized.

On the other hand, it is the need for testing. Even the technology provided by Leaf Book cannot guarantee 100% stability. If it is used internally, problems can be solved internally, which is more convenient to deal with.

There is another benefit that Ye Zishu didn't mention, that is, if most of the world's information runs on this cloud computing platform, it will be of great benefit to the cultivation of artificial intelligence.

Today's artificial intelligence can't do what science fiction does. Invading other people's computer systems is as simple as drinking water, and it won't be discovered.

Of course, if Ye Zishu personally trains artificial intelligence in hacking, he can still achieve a part of his ability, but it is impossible to intrude unconsciously.

But when other people's data is in their own fields, the situation is different. Artificial intelligence will have a large amount of information to learn, which is more beneficial than special training.

Specialized training and learning is just to enhance the ability of certain aspects of artificial intelligence, and this kind of overall training is to improve the overall level of artificial intelligence.

If you want special training before the project, you need to set the rules first, and then use the data for training, and the overall improvement is another operation.

Due to the vast amount of comprehensive data knowledge, artificial intelligence can learn from these data, increase its capabilities in various aspects, and then reverse optimize the rule settings.

That is to say, one is a process from theory to practice, and the other is to summarize theory from a large amount of practice, so as to continue to guide practice.

Although the former is more efficient and more specialized, the higher the certainty, the less surprising it will be. Everyone knows what will happen next.

Although the efficiency and specialization of the latter may not be as high as the former, but because there are no rules set in advance, it is completely artificial intelligence that learns and summarizes itself, and then rises to the stage of rules.

This will generate uncertainty and unexpected results. For example, after training and learning a large amount of mathematical knowledge, new mathematical theories or derived mathematical theories may be born.

Although such a theory does not necessarily have very high originality, it is also a summary of many mathematical knowledge and plays an important role in guiding the solution of some mathematical problems.

Of course, if the computing power of artificial intelligence is very strong, it is possible to produce highly original mathematical theoretical knowledge, and if it is popularized, this can also happen in other disciplines.

This means that artificial intelligence can already deeply participate in human scientific research and the exploration of the nature of the world. Perhaps in the future, human beings only need to provide knowledge clues and samples, and artificial intelligence is responsible for summarizing and sublimating.

Of course, these are theoretical possibilities, but in fact, limited by computing power, it is very difficult to achieve this, but it is not known whether quantum computers will bring about a qualitative improvement in artificial intelligence.

Quantum computers have much stronger computing power than traditional computers, but the calculation uncertainty is also much higher than that of traditional computers.

In layman's terms, the calculation result of a quantum computer is a probability number. The greater the probability, the closer to the answer, while the traditional computer is deterministic. This is the essential difference between the two computer architectures.

He really wants to know to what extent artificial intelligence can achieve after learning a lot of knowledge, and he is very looking forward to it.

As for whether strong artificial intelligence will appear, it is impossible to do it with the current computer architecture and software architecture. No matter how intelligent it is, it is just a tool.

And the function of a strongman is the same as life no matter how weak it is. It has its own thoughts, just like a baby. You can't deny the essential difference just because of your weak strength.

Others also expressed their opinions one after another, especially several technical directors of the company, who were still very interested in this and could find their own business points from it.

For example, Wu Chaoqiang and Fan Yongming of Xuanwu Technology Co., Ltd. think that the fully automated factory in the future can also be built on the cloud platform, which has many advantages over building an automated factory alone.

However, at that time, an industrial cloud platform may be built separately, and it will not be shared with the current cloud platform, because the requirements of the industrial cloud platform are higher, and many things are not completely consistent. This is also for security reasons.

So Wu Chaoqiang felt that it was necessary to obtain a piece of cloud computing-related technology. After returning home, he would study the industrial cloud computing platform and accumulate technology for the future.

The interest of Qinglong Technology Company is basically concentrated in the field of intelligent electronic equipment, not the kind of smart home mentioned above, which cannot give full play to the full advantages of artificial intelligence.

Qinglong Technology Co., Ltd. intends to combine artificial intelligence technology with electronic smart devices to empower more industries, such as smart cities, smart agriculture, and smart industrial applications.

Moreover, these fields are also an opportunity for Phoenix Technology Company and Xuanwu Technology Company. The three companies and even more companies can participate in it, and they can all obtain a share of the benefits that belong to the company.

Leaf Book did not mention these, but they can think of these through the above descriptions of Leaf Book, which makes Ye Shu have to feel that he only needs to release some cross-age technologies, and someone will naturally improve the application field.

Ye Shu naturally has no reason to object to their ambitious plans, although in his opinion, with their accumulated technology, there is still a long way to go to achieve this.

But in this process, many new technologies and applications will be born, which can continue to play a role, instead of having to wait for the end to create profits for the company.

At the end of this topic, Ye Zishu promised that he would take the time to sort out some technical materials and provide them with it, which he really couldn't refuse.

Because other people don't know how to do it at all, they just have more reverie from his description, and when it comes to the technical level, they have to be blind.