In the era of Earth, people often encountered a problem when training AI.
The quality of the data was too low, and at the same time, the total amount of data was too small.
For example, if people wanted to train an industrial AI, it obviously needed a large amount of real-world data from the factory floor.
But this data was very difficult to obtain. Why would a factory give you its data?
Even if it gave it to you, how could you confirm that all this data was real?
This was still when the situation was relatively good. Worse still were some AI training sessions that required direct data collection from the internet.
Data on the internet was mixed, and most of it could be considered garbage. How could it be possible to train highly intelligent AI with a pile of garbage?
But for Tom, none of this was an obstacle.
The data used to train this industrial AI set were all self-produced, all from the real factory floor. He could ensure 100% authenticity for every piece of data.
