I have discovered a relatively early new AI project called @ 4DLabs_Official, and I haven't tweeted much yet. At first, I saw YZi Labs investing and wanted to click on it to take a look. As a result, I became interested in its data business. I usually have a lot of graphics cards and memory, so I'm used to checking if the hardware is sufficient first. But there's still a bit of a problem with the robot. I bought the equipment, how can I teach it to work? A person takes a cup and reaches out to grab it. Robots need to learn distance and force, what to do if the cup slides, and how to grab it with a soft object. The small actions that we think are not worth mentioning, up to this point, are all lessons What 4D Labs wants to collect is this data. According to the design of the white paper, use first person glasses, tactile gloves and other devices to record real operations and process them into materials that robots can learn. There is also an entrance for scanning space and contributing scene data with a mobile phone. I quite like this. Data sources can enter different home and work scenarios. Similarly, when it comes to tidying up the table, there are probably ten thousand ways to deal with it. In reality, robots will have to face those chaotic situations sooner or later. Seeing this, I can better understand why it needs to be combined with Web3. Someone provides data, someone processes and verifies it, and downstream companies pay for it. Tokens are responsible for connecting these participants, so that those who contribute data can also receive rewards. If this matter can really run, the incentive mechanism of the cryptocurrency circle will have specific uses. The white paper also mentions that rewards will consider data quality, scarcity, and contribution to training. This one needs to be done well, after all, when it comes to brushing tasks, our circle has never lacked talent. People who are serious about collecting data should take more, so that someone is willing to continue working in the future. I am more interested in its future plan for the Model Store. After data collection and model training, the ability can also be sold to customers as a subscription, API, or deployment service. According to the design, these transactions will require $4D and will be accompanied by consumption and repurchase mechanisms. Speaking of which, if a company continues to pay, the token will have traceable business needs. Looking at the progress of the project later, there are also specific things to see - who is using it, how it is being used, and whether they will continue to pay. Of course, I am currently reading a white paper and have to wait for the product and customers to gradually submit their assignments. However, I think this approach is smooth and in line with my usual habit of looking at projects: first find a problem that someone really needs to solve, and then see how it can get the business done. The investment from YZi Labs made me click in, and after reading it, I am willing to continue following because it chose this incision. AI is moving towards reality, and someone has to do the delicate tasks of collection, organization, and verification. I have always been interested in this type of business. At present, the first round of Galxe Alpha activity has also been launched, and tasks such as X, Discord, Navigator Role, etc. can earn Alpha Badge/OAT according to the rules. If you are interested in this direction, you can go for a walk first and participate while watching the progress. https://app. (galxe.com)/quest/4Dlabs/GCJxRtZjhh