If you were correct, jetpacks and flying cars would be common. Kurzweil's main prediction is that human-level intelligence is essentially just a matter of having similar processing power to the human brain available, and then guessing when that will be based on current trends. It so happens that our current level of computing was possible with the amount of effort put in, but I don't think it was possible to predict that the amount of effort that could be devoted could produce the results we have, in the 1970s. In fact, I think the conventional wisdom of the time was that it was not possible to get to this level of computing power so quickly.
Strong AI is not a matter of "processing power". These pop-compu-sci guys are selling the 50s AI project which was already dead in the water upon invention (cf. Hubert Dreyfus' critique) but has since been realised by the compu/cog-sci people to be irrevocably flawed.
Today the thesis that Strong AI is a function of processing power is snake oil or a result of a serious lack of knowledge of this area. The latter is OK for a layman, not for this kind of "expert".
To be fair, Kurzweil has always said it will take about 10 years to develop the software for strong AI, after we have the hardware/processing power necessary.
True, however his prediction of computers being a million times larger and several times faster would be a massive improvement for current AI algorithms, not to mention what we will have by that time.
There were some pretty upbeat predictions for the future of computing in the 1970s - here is a quote from a contemporary review of The Mighty Micro from 1979:
"Some of his guesses seem based on other popular books: he thought 'ultra intelligent machines could prolong life to age 1000' - ie things which might measure blood or neoplasms, and even repair them."
