Clearly this would be an exciting announcement, but is Knuth even known to be working on the problem?

Only in his sleep.

Isn't every computer scientist supposed to work on P=NP in their sleep? Or maybe it's their lack of sleep.

I forget who it was that said it first, but one of my profs relayed it as serious advice: that every computer scientist should dable in P=NP in his early years, then go off and do something useful for a few decades, and come back to it to dable again.