That's actually not necessarily true. Forged carbon is virtually isotropic and can still outperform aluminium with the same shape (while weighing significantly less). If you actually design a part to make use of its lighter weight, you can end up with a part that's both stronger and lighter. It's just not something that you'd use for frames, because thin-walled tubes simply have better properties in both materials.I don't know anything about planes, but i do know that carbon fiber is strong in one direction, weak in another..
Once you make it strong in all directions, you end up with no weight savings vs aluminum
Most of the pics from the link you posted are cracks resulting from crashes, not regular riding. I am not going to deny that a carbon bike has less chance of surviving a crash with a car or overall big impact, but I also don't think it's something we should optimize for. And there's been plenty of metal frames that failed when landing jumps and the like, also exploding in a spectacular fashion, so those extreme situations are just something completely different than regular use.The most common failure is for the bike to suddenly shear in half while riding it without any warning signs.
Now I'm not a bicycle frame manufacturer, so I'm not sure where on the scale of strength vs weight they aim, but another property of carbon bikes is their massively improved comfort over aluminium. When done properly, a carbon frame will absorb vibrations much better while still being structurally sounder than equivalent aluminium frame. I imagine they would sacrifice stiffness deliberately for this effect.
Perhaps a really funny upshot of the latter are the newer Specialized frames. With the same geometry, the aluminium one has one pivot (around the horst link or even on that) more than the carbon one - because the carbon one can flex enough to not need a pivot. One less pivot, one less bearing, one less potential issue point.