I think the idea is that when a human checks to see if two hashes are equal, they usually just check the first few bytes, and maybe the last few bytes. You coul

I think the idea is that when a human checks to see if two hashes are equal, they usually just check the first few bytes, and maybe the last few bytes. You could attack this by grinding out hashes to find two with the same first few bytes. Of course, you could do something similar with this visual algorithm, but I guess the idea is you are using more of the human visual system so it's harder.