The objective of this invention is to provide continuous remote
authenticated operations for ensuring proper content processing and
management in remote untrusted computing environment. The method is based
on using a program that was hidden within the content protection program
at the remote untrusted computing environment, e.g., an end station. The
hidden program can be updated dynamically and it includes an inseparable
and interlocked functionality for generating a pseudo random sequence of
security signals. Only the media server that sends the content knows how
the pseudo-random sequence of security signals were generated; therefore,
the media server is able to check the validity of the security signals,
and thereby, verify the authenticity of the programs used to process
content at the remote untrusted computing environment. If the
verification operation fails, the media server will stop the transmission
of content to the remote untrusted computing environment.