Sir Martin Rees, Royal Society professor of astronomy at Cambridge University, agrees that it is now possible to conceive of computers so powerful that future societies may use them to build an entire virtual universe. In fact, like Nick Bostrom (who penned the first academic treatise on the subject), Rees wonders if we ourselves aren't the inhabitants of such a simulation.
These types of futuristic proclamations are nothing new for Rees, who has also argued that humans will eventually spawn machine intelligences and develop molecular assembling nanotechnology. It looks to me like he's keeping himself up to speed on all the transhumanist literature.
Now, on a related note, I'd like to take issue with a claim made by Jonathan Leek, the author of the Times Online article about Rees (which is really quite weak). In the article, "Top scientist asks: is life all just a dream?", Leek writes that René Descartes "famously came down in favour of existence, saying: 'I think, therefore I am.'" He uses this as possible evidence in favour of the argument that we are not in a simulation.
This is not an accurate interpretation of Descartes's assertion, nor does it disprove the Simulation Argument. In fact, Descartes's dictum opened the door to the possibility that existence as it is perceived may be a complete fabrication of some sort. Cartesian skepticism suggests that the only thing we can know for certain is that we can be aware of our own conscious, subjective self.
Further, a simulated reality would still constitute a bona fide 'existence' that could be experienced by a self-referential agent. I believe Leek is suggesting the contrary. A simulated existence is still an existence so long as it is perceived by a conscious observer and a high degree of modal realism is maintained.