Scott Aaronson expresses his confidence that the proof is wrong by offering to supplement the million-dollar Clay prize with $200,000 of his own money if it's correct:
I suppose he could have safely claimed this with Wiles's original proof as well, since it did have at least one non-trivial flaw that required amending. It's quite possible that the proof contains flaws, but will still hold up in the end, because the flaws can be corrected. But I feel that's a bit of a lame gamble.
But I cannot make a qualified guess if this might be the case here.
> If P≠NP has indeed been proved, my life will change so dramatically that having to pay $200,000 will be the least of it. […] If P≠NP is proved, then to whatever extent theoretical computer science continues to exist at all, it will have a very different character.
and
> I can afford $200k, but not in the same way Bill Gates can afford $200k.
http://news.ycombinator.com/item?id=1587405
As randomwalker says, it's not entirely obvious that Aaronson is betting against the proof.