I was under the impression what made glass fragile was not it's hardness or elastic modulus, which are already very good, but crack propagation, which is exponential.

Glass is harder than steel and has a comparable elastic modulus to many metals, the problem is that any tiny fracture becomes a multiplier for stress so once it starts to go it goes very quickly at the point of any imperfection. Glass with a perfect surface can already withstand phenomenal amounts of stress, until it eventually goes off to powder like a prince Rupert's drop instead of breaking normally into shards.

Making a corundum/glass hybrid is all well and good but I don't think this will replace steel. Steel likes the little cracks that form in it, they reduce internal stresses and then don't get any bigger.

That's interesting. I don't think it will replace steel, but having a "tougher" glass means thinner glass in for instance aerospace applications where weight costs money.

I think if this is a mixture of glass and corundum, it could prevent crack propagation. As the mixture cools, you are going to have corundum crystallizing at the interface with the glass. If the glass cracks, it would propagate through itself until it reaches the crystal boundary of the corundum and stop if it doesn't have the strain to break it (since it doesn't have cleavage and has a Mohs hardness of 9)

How is this different from ALON (aluminum oxynitride). Is it just supposed to be cheaper since it's mixed with glass, or eventually easier to make?

http://www.forbes.com/sites/carmendrahl/2015/11/03/sorry-but-that-new-glass-from-japanese-researchers-isnt-unbreakable/