I believe that we will eventually have an AI that will meet most observers' definition of consciousness. That being said, I think there existed serious people

I believe that we will eventually have an AI that will meet most observers' definition of consciousness. That being said, I think there existed serious people who thought we would have one before 2020, even as late as 5-10 years ago, using technologies that don't seem wholly adequate to the task now (expert systems, fuzzy rule systems, multi-layer perceptrons, etc). At what point does the non-existence of machine consciousness begin to make us skeptical that such a thing is possible, that no algorithm describes consciousness?