A recent slashdot post links to this Wired story and asks the following oft-asked question: Will AJAX Threaten Windows Desktop? BTW, there is nothing new in this, as it has been opined before in a copule of well cited writings: How Microsoft Lost the API War by Joel Spolsky and The Location Field is The New Command Line by John Gruber.
Comping back to the Slashdot post -- despite the general animosity towards Microsoft among Slashdot readers, most commenters don't equate emergence of AJAX (and its perceived promise!) with death of Windows API, at least not yet! Quite rightly so, IMHO.
In my opinion, AJAX today is very much like Linux in 1995 -- full of potential, lots of momentum and unquestioned acceptance among early adopters. I have no doubt that wide availability of AJAX frameworks will change the way Web Apps are written. And that would be a good thing. We have all used intranet Web Apps that suck big time, cause lost productivity due to page loading wait time, and unwieldy interaction with the user. Use of AJAX will certainly make these Web Apps more pleasing and responsive, simply by moving a lot of processing to the client machines and replacing reload of pages with more localized cahnges (to the page, or more precisely, to the DOM tree underlying the web page).
But keep in mind that early Linux enthusiasts were those who were already familiar with some flavor of Unix, and not those whose only encounter with PCs was through DOS or Windows. Something similar is likely to happen with AJAX. It is exciting for folks who have been hamstrung by the current Web App paradigm in delivering useful functionality and are looking for alternatives to build better Web Apps.
In fact even now, Linux is not hurting Windows as much as the various flavors of Unix. And that brings to my conclusions about who is going to get hurt most by AJAX , at least in the short run -- It will be Java applets, Java Swing and Java WebStart. Java applets are not much in use anyway. Swing is cross platform, has seen continued performance improvements (partly due to JVM optimization and partly due to Swing class library optimizations) and has the potential of distribution through Web using Java WebStart. Due to these attributes it has seen some success in certain quarters lately. However, AJAX will challenge the choice of Swing for such Apps. I wouldn't be surprised if more and more development teams lean toward AJAX than Java WebStart. There are still some hurdles -- such as ability access client machine resources, but these will be addressed in near future.
However, in the long term, AJAX could be used to build more and more applications replacing Windows APIs, especially as people become comfortable with the idea of using hosted applications and accessing them through browsers. The same way as Linux is making slow but steady inroads into Windows user base!