First of all, I don't like the 'turing test' because fooling a human doesn't imply intelligence, survival should be the evaluation. The next best thing to test is Lossless Compression (because testing survival requires watching long term, inventions...). Of course copying humans will lead to better prediction, same thing; turing test. But use Lossless Compression to make it sound human, and then finally don't give it the turing test because that IS the test; Lossless Compression. You don't need to test the predictor after that.
So yes and no, yes the visual turing test would be to draw the correct human-like entailments that follow or that are similar semantically (or segmentation), aka good predictor, and no we wouldn't do any visual turing test on it, simply we would train a visual lossless compressor to make sure it compresses it most and can regenerate it back perfectly and be done after that, no further testing. The data to compress visually would be human knowledge in videos from Wikipedia :-) ------------------------------------------ Artificial General Intelligence List: AGI Permalink: https://agi.topicbox.com/groups/agi/Tc67faac3048278cf-Mc212656d5cffcfc6ce298dae Delivery options: https://agi.topicbox.com/groups/agi/subscription
