No, I'm afraid the auto-vectorizer doesn't use AVX in this first release. (But high up on our TODO list!)
In VS2010, the /arch:SSE and /arch:SSE2 switches, on x86, tell the compiler to use XXM registers for floating-point calculations, rather than x87 registers. Similarly, the /arch:AVX switch tells the compiler to emit AVX instructions, rather than SSE instructions. But in all those cases, it emits just SCALAR instructions. It is only with the advent of the auto-vectorizer that the compiler emits SIMD instructions that make full use of the wide XMM vector registers.
*AX, *BX. Yes, sorry. No excuse! (As I replied to Bruce, above, I was more careful with this example in the blog)