Hm... this is a fairly complicated issue.
It is somewhat true that (some of!) the antiphishing databases AOS uses use newer data. This is mostly due to 2 main reasons:
1. We tend to include only "used" phishing in our VPS. That means that if a phishing page was not visited within our userbase, we do not add it to VPS. Doing otherwise would cause our program to take up more space (both on drive and in memory) as well as take up more bandwidth (bigger streaming updates).
2. We tend to include only "verified" phishing in our VPS. We do not fear including potentially false positive URL in AOS (our browser plugin), because there are less things to break in case of a false positive (keeping in mind that the browser plugin can only interfere with browsing), but analysis takes time and until we are sure that it will not break anything when it will be scanned for in all your hard drive, memory and network, we do not want to add it to VPS.
If the phishing URL satisfies both points, there is a very stong likelyhood that it will be added to VPS (and, therefore, webshield).
Of course we are constantly making sure that there is as little delay as possible
!
