Yay! Waddy's decided to post one of those "what's wrong with this picture" puzzles. Let's play!
1) In the motherboard photo, there's obvious unblended sections everywhere. The most prominent is above the left socket, across the white diagonal line to the left of the vertical orange trace. There's a quite apparent joint (at 160x, if you happen to have an image ruler handy) that cleanly runs straight down the y-axis, in spite of the image being tilted slightly clockwise. If the image were more compressed than it is, it could MAYBE be written off as jpeg banding, but there's no evidence of jpeg artifacts anywhere else in the photo.
2) As we look at the rear of the chip(s), everything looks quite kosher -- until we properly scale and rotate the image and transparently place it on top of our socket. My gosh, AMD has made a chip that doesn't fit! Won't their engineers be embarrassed. The spacing between the two sets of pins is narrower than the sets of holes in the socket.
3) If you're going to build a double-chip, it's assumed that both chips are the same speed, right? Then why do we have TWO SETS OF BRIDGES? Redundancy isn't something that often comes to hardware (with the possible exception of prototypes -- and if this were a prototype, it certainly wouldn't be built in the Malaysia plant, much less have the fact advertised.) Also, this chip has the remarkable ability to bend light, as the shadows under the HSF pads are duplicated for each side of the chip, with the "middle" pad apparently taken from the upper-left. (The original image apparently showed a centre-right light source, so the shadows across the top pads would actually be falling gradually to the left, instead of centre-left-centre-left as shown in the photo.)
One could argue as well that a design such as this would be EXTREMELY inefficient and prone to burn out, since the units would be transmitting heat between one another before it ever got a chance to be dissipated by the heat sink. This is thermal engineering at its very worst. However, I'm not an engineer, so this is the one point that's simply conjecture.
I don't wanna make a fuss, but the frequency of these sorts of bogus, unverified "inside info" news postings seems to be increasing at quite a rate lately. I know it's a difficult task to verify the horde of submissions that come across Waddy's virtual desk, but as an early-on contributor to xp.org, I'd hate to see this respected news site denegrate into a tech tabloid.