In the new book BOLD by Peter Diamandis, Kurzweil reveals his predictions for the next 25 years. It's worth noting that an equivalent set of predictions, offered 25 years ago in 1990, turned out to be startlingly accurate (a computer would defeat the human chess champion by 1998, PC's would routinely operate over wireless networks by 2010, etc.).
By the 2020s, most diseases will go away as nanobots become smarter than current medical technology. Normal human eating can be replaced by nanosystems. The Turing test begins to be passable. Self-driving cars begin to take over the roads, and people won’t be allowed to drive on highways.
By the 2030s, virtual reality will begin to feel 100% real. We will be able to upload our mind / consciousness by the end of the decade.
By the 2040s, non-biological intelligence will be a billion times more capable than biological intelligence (a.k.a. us). Nanotech foglets will be able to make food out of thin air and create any object in physical world at a whim.
By 2045, we will multiply our intelligence a billionfold by linking wirelessly from our neocortex to a synthetic neocortex in the cloud.
The meta-theme of Kurzweil's predictions is that of exponentiality, extremely rapid growth (e.g., in semiconductor density, computing power, ...) in the overarching capabilities of man's inventions.
Which makes me think of Skynet.
As the wise man said, "May you live in interesting times."
Hat tip: BadBlue Tech News.