"Or, most likely, it will develop a model of the data it was trained on, and not the world this data came from." I mean the point is what is the difference? Is

"Or, most likely, it will develop a model of the data it was trained on, and not the world this data came from." I mean the point is what is the difference? Is your brain trained on 'the real world', or on your senses (data) of the real world? Descartes argued that you can't tell if there's a real world at all. By 'real-world sensor data' I mean that the models can be trained on sensor data from the real world (video, audio, feedback from motor outputs, etc), rather than on text, which is only abstractly related to the real world. I believe this is called the token-grounding problem, and goes back to Searle's Chinese Room thought experiment.