Matrix April AMA Transcript

Matrix AI Network
18 min readMay 3, 2023

--

Eric: Hello, Matrixians! Good morning,good afternoon and good evening.

So it’s time that we run the AMA for the April. Today is 27th of April. So as usual, we have our CEO, Mr. Owen TAO to join this AMA. And this time, just like the March one, we will conduct this AMA in Chinese.

So Owen can speak more friendly with more details, so that you can get more insights from him and understand more about what we are doing and why we have been doing like that. So without further ado, let’s start with the first question.

The first question is from CryptoAlex7. Now we turn back to the language of Chinese. The first question is from CryptoAlex and it is about Morpheus. TAO, please tell us when can we test our Morpheus? Because everyone is looking forward to our own version of ChatGPT.

Of course, our Morpheus is not for ChatGPT to compete with it. It is for our own purpose in our 3.0. So please, Mr. TAO, tell us about the Morpheus test.

Thank you, Mr. TAO.

Owen: OK, thank you, Eric. First of all, we design Morpheus not to be a tool like ChatGPT. In our 3.0, Morpheus will have some important functions. For example, when we generate this avatar intelligence, in addition to the brainwaves, we also need a tool based on a large language model to collect some of the user’s personal thinking and combine them with the brainwaves. In addition, we can also use it in the future to generate some personal virtual characters in our 3.0, to enrich the world of the universe. This is one of the original intentions of our design of Morpheus.

We are still doing some internal tests and adjustments. As you can see, we are already collecting volunteers from the community to participate in our tests. So yes, we have a plan to launch our test version. But we are still working on some adjustments. I expect to finish it in one or two weeks and open it to the user. But we don’t have as much support as OpenAI for computing power. So we will definitely have some number of people for our user tests. We will adjust, optimize and upgrade it later. This is our current plan.

Eric: So we will launch the test in two weeks. We will also adjust the number of people. We will announce it to the community through an announcement. This is the first question.

The second question is from Angel Maria. This question is about what Matrix can do in social media or online communities to solve some problems. Please go ahead, Mr. TAO.

Owen: Okay. First of all, the main contribution of Matrix to the future of social media or online communities is probably from our 3.0 part. First of all, whether it is Web3 or the original universe, we cannot live in the virtual world all the time. So Web3 or the original universe has its own social media. Our avatar intelligence can replace us to process some information or generate some content.

Secondly, this is also what we have been thinking about in the recent Morpheus test. Whether it is the original universe or Web3, it has a big difference from the physical world. In the physical world, we have 7 billion people in the physical world. In the virtual world, the difference between the virtual world and the physical world is that we have 7 billion people in the physical world. At the same time, we have a border in the physical world. We are as big as the Earth. We have some space in the ocean. The land is for humans to use. The biggest difference between the virtual world and the physical world is that people in the real world cannot be online all the time.

The second difference is that the virtual world has an infinite space to expand. So from this perspective, the virtual world is much more open than the real world. So, in addition to the users of the physical world, it needs a lot of artificial intelligence or robots to play various roles and provide us with various services to fill this world. Normally, robots in games cannot fully meet our needs. So we need some robots with independent thinking or personality. At the same time, we need robots with strong reactions and interaction. This is what we thought about when we were working on Morpheus and 3.0. This is still under exploration. In addition to creating our own avatar intelligence, we also have the opportunity to create third-party ones.

For example, I can use my favorite actor, for example, I like Julia Roberts. I can use her image and personality to make some adjustments to my personal preferences to train some digital humans. In the future, there will be a lot of digital humans to fill the space in the metaverse or the outside world. I think this is very valuable. This virtual world will be more possible and more interesting. This is what we are exploring now. I have talked about online communities before. The most important thing in online communities is the common Daoist organization. We can’t participate in all kinds of Daoist activities in real time. If we can use a way of thinking that is consistent with our values, a way of thinking that is consistent with our values, we can deal with a lot of things that may not be so important.

We can also make the governance of the entire community more orderly and form a decentralized community that is not limited to the Daoist community. This is what we are exploring now and what we may be able to do in the future in this field. Thank you.

Eric: Mr. TAO, you mentioned something interesting. How can we use Morpheus or 3.0 to make social media more interesting? This is a very interesting point. You mentioned that you can learn from other celebrities or certain people. You can learn from them. This is similar to what we are doing now.

For example, ChatGPT. When you enter the prompt, you can also have a similar function. You can use ChatGPT as an AI scientist or a chemistry teacher or a certain role.

This is also what we express in text. We can also use our own Avatar Intelligence, our AVI, and then use Morpheus to realize this interesting avatar game or a very independent person who has been trained. This is different from text. It is a real avatar in a web3 world to help you to show more interesting features. This is what Mr. TAO just shared with us about some contributions in social media or some new and interesting ideas. Thank you, Owen.

Next, let’s move on to the third question. It’s from redbit4234. How can Matrix solve potential problems including data ownership and governance? How can we ensure that users have their own data in their own hands? Owen, please.

Owen: Let me think about it. Let’s divide this into three points. First, we have realized most of the functions in the 2.0 version. First, we need to solve the problem of data ownership and governance. This is a very simple function for the blockchain. When we upload data, the blockchain will identify the data ownership by means of a time loop. In fact, when we trace the data ownership in the future, we can do it in the earliest principle. The data ownership of the first person to upload the data is the first person’s. This solves the problem of data ownership and governance.

Second, in order to avoid the easy copy of data, for example, if we store the data on Alibaba or AWS, the engineers of Alibaba or Amazon can see all of our data. This is a hazard. They can easily copy our data. So, after we solve the issue of data ownership, the next most important function is to store the data in a distributed way. Now, Matrix has deployed our own IPFS network.

We hope to find a more mature third-party network to support our data storage in the future. After all, professionals do professional things. For example, Filecoin is not yet able to meet our needs. So, we have been using our own IPFS network. After our data is uploaded to the IPFS network, each node only has a small part of the data. It can’t see all of the data. So, it can’t use the data. So, we avoid the risk of the data being copied. The last and most important point is to separate data ownership and usage. To achieve this goal, we need two very important technologies.

The first technology is distributed computing. Yes, MANTA has already realized this function. By distributed computing, for example, if Eric wants to train an AI model with my data, then my data will be divided into many parts and used simultaneously on different machines. For Eric, he can’t get a complete copy of my data. So, Eric can get a complete result but not my data.

This is the first point. It is very important to use a technology that uses distributed computing. Above this, we need a more important technology that uses privacy computing. That is, when each node is computing with our data, it can’t see my data. Although my data is divided into many parts and computed simultaneously on many nodes, the more perfect solution is that in addition to distributed computing, we need to do a layer of encryption.

That is, when each node is computing with my data, it can’t see the complete data. We haven’t achieved this yet. We will gradually explore this in the future to improve this function. From this perspective, we can ensure that the user’s data is first, authenticated on the blockchain and is safely stored in distributed storage. At the same time, we can use distributed computing to separate the user’s rights and rights of use. The third point is that we need to separate the user’s rights of use from the rights of all users.

Now we have adopted the distributed computing of MANTA. Then we can separate the user’s rights of use from the rights of all users’ data. In the future, we may also want to upgrade and add a technology similar to encrypted computing to further protect the user’s data. These are some of the work we have done and some prospects for the future. Thank you.

Eric: Okay, thank you, Mr. TAO. The next question is from Hub21. What is the long-term goal of our MATRIX? How do we measure whether we have succeeded?

Owen: Our ultimate goal is to establish the Matrix project from the very beginning. Of course, we were also affected by the movie Matrix. We hope that one day we can realize the scene in the movie. So, from our long-term goal, the movie Matrix is our ultimate goal. In the short term, we have established a project called Matrix 3.0.

This project will be divided into several stages. We will also work around this goal. This is what we have said in our official website and in the previous articles. We will work around this goal step by step. The first step we will take in the short term is to build the most basic infrastructure for our brainwaves.

The next step is to work around these basic facilities to further improve the quality of our data and provide more applications. The next step is to see whether we can find more data and deeper applications in our brainwaves. Then we will establish a division in the virtual world and move forward step by step to the stage of the movie Matrix. In the process, we will also work on the data and we may find some new applications that we didn’t expect when we first designed the map.

We have found some new applications. In June or July, we may introduce and show some new features that are different from the original map. How do we evaluate whether we can successfully influence the movie? I think there are two key points.

The first is whether our product is good. Can it really solve the problem? Can it really help us to have a different experiencein the virtual world? The second point is a more important indicator. How many people will use our products and services? In the future, as we move forward, we will show you more in-depth information in the form of tests and commercialization. We all think that good products are not used by many people. But it is a failure. I would like to share a few key points with you.

Eric: The first is that as we explore the Matrix 3.0 more deeply, and as we continue to expand our technology, we will make some corrections to the roadmap.

We will make the road map more textured and more grounded. So we will follow the progress of the roadmap. We will also correct these roadmap to make it more interesting. It can be better in the future. In June, we will make some adjustments to the first stage of delivery. We will tell you more about this in the announcement. In addition, as Mr. TAO mentioned, the next question is about how we can prevent the risk of brain waves. Mr. TAO, please explain.

Owen: Okay. As of now, we have not commercialized it. We have already created some applications for the brain waves. Let me explain. First, in terms of medicine, we are currently trying to diagnose and treat the patients with depression. In fact, in traditional ways, we evaluate patients through questions and observations. In the new system, we have added the evaluation of the brain waves.

When a patient with depression answers a question, if he says something false, we can use the brain electric waves to judge whether he is excited about something. This can be used to evaluate his treatment. We can also effectively diagnose patients with depression. In addition, in terms of education, it is related to medicine. In terms of children with a lack of attention, we can find out early on whether their attention is deficient. This is actually shown on the brain waves. It is visible. At the same time, we can give them some targeted guidance and treatment for the waveform of the brain waves.

This is something we can do in the medical field. As for the game, we are also exploring. We are currently trying to realize some scenarios in the game. I think we will see some details in June and July when we show the project. For example, we can use the brain electric waves to control the speed of a racing car with our attention. Can it go straight? This is something we can do at the moment.

However, just using the brain electric waves to realize a game effect is not a very good thing. You can’t feel a very good game fun in this. What it can do is to train your attention and your brain activity based on the current game. This is not a very interesting thing. To solve this problem, we have made some expansions. We have some set of buttons. Of course, this is a bit different from our 3.0. We found that in addition to collecting brain electricity, we can also collect the brain’s energy.

We found that in addition to collecting brain’s energy, we can also collect the brain’s energy. For example, when I lift something, I can clearly see which muscle is exerting force. What is the degree of force? What is this application scenario? Now, people will play games like Microsoft’s Kinect. Or the Wii’s physical game that is popular on the Internet. Most of the physical games on the Internet are made by capturing your actions through a camera.

There are some problems with that. For example, it can’t accurately assess your actions. At the same time, when many people are playing the game at the same time, it will be very chaotic. It can’t support many people playing at the same time. From a machine learning perspective, it can be used to expand the game’s physical sense. For example, it can accurately assess everyone’s actions. It can clearly know which muscle is stretching or contracting.

What is its degree? This is very clear. At the same time, it can support many people playing at the same time. I believe that if we want to do this, or if a game company is interested in cooperating with us, I think this will become a very interesting system for the physical game in the future. At the same time, in the future, in the distant universe, in addition to the brainwaves, we will add our machine power. Users will have a new experience and perception of the distant universe. It will be completely different from VR. These are our thoughts. As for how to prevent brainwaves, there is something very important we are doing. We will probably show you something in the future.

We are working hard to find some unique factors in brainwaves. We want to make brainwaves similar to fingerprints and to be able to be certified by biological information. When we have something like this, the brainwaves will be able to detect and capture the brainwaves easily. For example, some fingerprints on the market can be copied from other people’s fingerprints. It is a bit of a hassle to collect genes.

I think brainwaves will be a good, unique biological information recognition. If we can do this, I think we can prevent the so-called brainwaves from being loopholed to a large extent. This is what we are doing now.

Eric: Thank you, Mr. TAO. You answered in great detail. The next question is about… Okay. The next question… We talked about the brainwaves. The next question… Some people are interested in developing on our platform. Can we provide some online courses or activities to let them know how to… Okay. Mr. TAO, you can answer this question.

Owen: Thank you. This is a very good suggestion. We have some development documents and some content related to our project on the main page, including some documents on the platform. Of course, for many people, this content is not detailed enough. Based on your suggestion, we might consider organizing our own online courses at the right time to show you some explanations and demonstrations for each part of the video.

Just like what we are doing now on the MANTA deployment and the video. These videos will be uploaded to our website in the future. You can get some more detailed information about the operations and guidance. If you have any additional questions, we will organize them. If there are many typical questions, we will record new videos or make new guidance to help you use our tools and deploy new applications and developments.

Eric: Okay. Just now, Mr. TAO mentioned the Manta video. Let me give you an example. This is also a case for development in the technical field. We have a place called Resources on the website. You should be there. This is the information you want to find.

Of course, if you have any other needs, we will make some explanations in the form of videos. If you have any technical questions, we can also hold a so-called tech talk to explain these technical issues. In these technical sessions, our technical staff are all Chinese, so we will have a Chinese-Chinese explanation session. We will add subtitles later. This is another way to do a tech talk for a specific topic.

This is a question from EB Magic. The next question is from Bright Patrick. He has a proposal that we are now in two centralized exchanges, including KuCoin and MEXC. He has a proposal to provide a collateral to these exchanges to look at this issue. Mr. TAO, what do you think about this? Do you want to share it with Bright Patrick?

Owen: Okay, this is a very good suggestion. We will also communicate with these two exchanges to see if they have made similar cases in the past and how they did it. We will also see if it meets our needs. Before that, if you want to make some staking, you can choose to establish or participate in our verifier node.

This is actually quite similar to some staking methods. As a verifier node, you need to deposit some of our money in it and you will also get some of the profits. Of course, we will also promote your proposal. We will discuss with KuCoin and MEXC to see if we can work with them. Thank you.

Eric: Okay. Bright Patrick. I will deal with this issue. I will first go to KuCoin to see how we can work together and what they need from us. We will continue to follow up on this. Mr. TAO just answered the question from Bright Patrick.

Next is the last question from our AMA. This question is also from EBMagic. There are several questions. One is… What are the current applications that MANTA is running? What are the upcoming applications that MANTA is running? This is the first question. The second question is what are the current applications that MANTA is running? What are the next applications that MANTA is running? This is a question from EBMagic. Mr. TAO, do you have anything to share with EBMagic?

Owen: Okay. Thank you. First of all, MANTA is currently running the AI services that MANAS is currently using. The computing power is provided by MANTA. At the same time, we have added some new features to MANTA, including some automatic machine learning deployments for image recognition.

Yes, we are waiting for MANTA to release guide video so that you can see it. We have also found some partners for this feature, including some universities and companies that need machine vision. Yes, these are some of the scenarios that MANTA is currently using. In the future, we will use MANTA to run some of our future operations, including some of our 3.0 applications.

Our Morpheus is already deployed on MANTA but we have not yet opened it for everyone to test. I believe you will be able to see the new product soon. This is the future situation that MANAS will be using. MANAS currently has all kinds of AI programs on it. All kinds of applications, including some of the recognition of some of the landmark buildings, including face recognition, and so on.

In the future, we will make MANAS an open platform.In addition to sharing some of our results, we also welcome scientists from all over the world to share their good news about their AI applications or some of their features on MANAS. MANAS will be an open platform. Anyone can use the MANAS API to build their own AI service platform or such a… Economic model.

Yes, or integrate it into their own application. The scientists who provide the service will benefit from the number of services that others use. At the same time, this is also beneficial for MANTA. All these services are provided based on MANTA’s algorithm.

So, if someone uses MANAS service and your node is selected to provide the algorithm for the service, you will get a return. In the short term, we will make some adjustments to MANTA’s operating strategy.

Due to the recent expansion of the large-scale explosion-style AIGC, MANTA will not focus on all areas to provide a universal solution. We may focus on AIGC. We are also trying to deploy Morpheus on MANTA. So, in the short term, MANTA may focus on AIGC and hope to provide more growth for MANTA with the growth of AIGC. These are our future plans. AIGC is very popular now.

Eric: I think this is the right direction and a very promising thing. Now everyone is discussing 3.5, 4, and even 5. There is also about the mid-yearly stable diffusion. AIGC is definitely a trend of growth globally. We will make some adjustments to the strategy.

Before we end this AMA, we mentioned that we will make some adjustments to the distribution of our new year’s Eve products. Can you tell us what kind of adjustments you will make?

Owen: Sure. In June, we will make adjustments to 3.0. I already mentioned that we will make some adjustments to the future scenarios. We won’t push it to a universal way. We will focus on AIGC in the short term. This is the first adjustment. The second adjustment will be in 3.0. We won’t make a universal one in June.

We will focus on the display and platform of all kinds of brain chip data. We have found that it is difficult to reach this goal in the past few months. We are still exploring. So, in June, we will focus on the chip of our partner’s brain interface and make some applications that correspond to it with this standard and data.

Of course, in June, we can show you some applications that use the brainwave. We will also show you some that use the brainwave. For example, we hope to show you some of the applications that use the brainwave. In addition, I mentioned that in addition to the brainwave, we may have some opportunities in the future for muscle wave. We hope to show you some of them in June and July. You can also see if there are any good applications in the market.

Eric: I think these adjustments are like the behavior of the market. There are some so-called adjustments. This is a reasonable thing. Especially when we go deeper and actually develop these developments, we will understand what direction we should go in the future.

We will also make some adjustments to the world trend. These adjustments are to make the whole project better or more grounded. In June, we will make an announcement and we will also show you some of the things we will show you in June or July. We will list them in detail.

This is our AMA in April. This is the end of the AMA. Thank you, Mr. TAO, for sharing so much information with us. I believe that our Matrixians can also help us to understand our project and some of the ideas and adjustments. We can also understand why we do this. This is the end of the AMA. Thank you, Mr. TAO.

Owen: Thank you, Eric. Bye-bye, community.

The Matrix AI Network was founded in 2017. In 2023, we enter Matrix 3.0 blending neuroscience with our previous work to realize the vision of the Matrix films.

FOLLOW MATRIX:

Website | GitHub | Twitter | YouTube

Telegram (Official) | Telegram (Masternodes)

Owen Tao (CEO) | Steve Deng (Chief AI Scientist) | Eric Choy (CMTO)

--

--

Matrix AI Network

The Matrix AI Network was founded in 2017. In 2023, we enter Matrix 3.0 blending neuroscience with our previous work to realize the vision of the Matrix films.