Very good point. Isn't this kind of innovation restricted to the Kittygrams that operate at large scales, though? Kittygram will need to reach a large mass of

Very good point. Isn't this kind of innovation restricted to the Kittygrams that operate at large scales, though? Kittygram will need to reach a large mass of users, get bought out, face the need to extract specialized data from images, and finally hire expert engineers to work on that problem before they can innovate. How many startups get to that point? Even if the Kittygram founder had a vision of building an innovative object recognition solution and merely used Kittygram as the means to an end, it feels as if there must be a more efficient way to solve these problems.