Ultimately the main benefits to using claims include:
providing a consistent programming model for your services - you
don't need to know how to implement a particular security mechanism,
one site might use username and password
authentication/authorisation, another Active Directory. You services
don't care either way because all you are doing is processing the
claims in all instances.
You don't need to concern your self with the security
implementation. This is done by a third party.
you can customise claims to suit your domain, and treat them as an
extension to your authorisation logic - standard security properties
usually only provide you only with basic information such as roles.
You can of course extend this but then your doing much more work and
is often difficult to implement (eg. extending AD is often not so much a technical challenge but a policy constraint - admins are reluctant to modify the AD schema to accomodate a specifioc application).
Interopable - because the claims [format] are based on standards they become much more interopable between services of different languages and domains as the underlying technology for the security is abstracted.
If you are creating new .NET 4.5 WCF Services you can already start using claims as the namespace is backwards compatible with earlier security implementations, so even if you did decide claims wasn't for you now, you would be in a better position to upgrade later.
There is much more to claims than I can write here and I'm sure there will be others with additionbal reasons why considering claims might be a good thing.
Hope this helps.