I installed Office 2003 Service Pack 2 and the latest junk e-mail filters for Outlook. The process was painless. And the anti-phishing features are interesting.
Details about bug fixes are in this KB article. After a five-minute search I can’t find any documentation of how the new anti-phishing features work. But these are my observations:
- All messages that appear to be phishing attempts are moved to the Junk E-mail folder.
- All HTML-formatted messages in the Junk E-mail folder are displayed in plain text. This is a crucial change, because it denies the scammer the opportunity to steal the look and feel of a legitimate site. Even if the scammer tries to steal a site’s graphic, the effort is in vain, because all you see is a link to the graphic.
- Links are broken up into the link text and the link target, which appears in brackets. As this screen snippet shows, it’s pretty easy to spot the phony links. As a bonus, the link text is not clickable. You have to copy the URL and paste it into a browser’s Address bar to actually visit the site.

The forced conversion to plain text also renders a lot of spam unreadable, which is good. So-called online drugstores that try to disguise their content by burying the message text in a bunch of pseudo HTML just turn into so much gibberish.
What if the junk/phishing filters catch a legitimate message by mistake? No problem. Drag it back into the Inbox or any other folder and it’s displayed in its original format, complete with clickable links.
This is a simple but very effective fix. If you use Outlook 2003, go get it!
Update: Thanks to Rick in the comments for finding this link to Microsoft’s brand-new Help topic: Block or unblock links in suspicious phishing messages. In addition to the features I noted above, there’s a new link-blocking behavior that applies to messages that contain suspicious links but aren’t moved to the Junk E-mail folder. Here’s a screen from the Help topic:

Unlike the spam filtering, this classification isn’t retroactive; it applies only to new messages as they’re received. So I won’t be able to see it in action (and show it to you) until I receive a new, suspicious phishing attempt that doesn’t get classified as spam. We’ll see how long that takes.

