I am having this same problem - "Safari has encountered a problem and needs to close" blah blah
I agree with Federico Parodi, it looks very like the problem with iTunes 22.214.171.124, which has now been fixed.
That crash was apparently caused by "an SSE instruction set that is not present on some older CPUs". For some reason, Apple do not have any hardware that matches this configuration in their QA lab. Any Apple techs looking might wanna check that out.
Only difference I notice is that, for me, Safari looks as though it's about to load, then crashes. With the iTunes crash, it went straight to the error with no delay. I just see the top of the Safari window, then it crashes. Still, it looks similar.
Oh well - firefox is probably better anyway.
PC Windows XP
I have the same issue. In my application log I see the following:
Log Name: Application
Source: Application Error
Date: 6/11/2007 2:19:47 PM
Event ID: 1000
Task Category: (100)
Faulting application Safari.exe, version 3.522.11.3, time stamp 0x466c3cec, faulting module CFNetwork.dll, version 126.96.36.199, time stamp 0x466c263f, exception code 0xc0000005, fault offset 0x0000af81, process id 0x155c, application start time 0x01c7ac6e3374fb8e.
<Provider Name="Application Error" />
<TimeCreated SystemTime="2007-06-11T21:19:47.000Z" />
Dell D620 Windows Vista
Indeed, this does appear to be another case of the dreaded SSE instruction bug, at least from what I can tell.
The exception code in the crash report that I'm encountering is 0xc000001d (illegal instruction) - just like in the iTunes bug.
The exception occurs in Virtual PC for PowerPC Mac (which has no support for SSE instructions), but I don't get the error on a real Pentium 3.
PowerBook G4 1.6GHz Mac OS X (10.4.3)