The original 2017/18 experiments by @notmisha @brandondamos @serkancabi @sergomezcol @laurent_dinh @yuvaltassa and a few other colleagues.

@NandoDF: A blind robot that predicts what it expects to sense next with its touch sensors, learns an internal representation of what the world looks like. In this case, predicting hand sensors enables to robot to understand the orientation of green cake slice.

I believe this accentuates the need for us to move beyond text, audio, and video into touch, vestibular information and proprioception. Self-awareness is likely to be an emergent property arising from predicting multimodal streams of data. This is an exciting research hypothesis that needs to be probed further with scalable experimentation.