source: http://stackoverflow.com/questions/311064/are-there-public-key-cryptography-algorithms-that-are-provably-np-hard-to-defeat

They are outside of P but are not actually hard for NP. Therefore proving P=NP is not useful in solving these problems.

Not to burst anyone's bubble, but the factoring problem is actually not known to be NP completely and evidence points to the fact that it is not, since if it could be shown to be then this would prove that NP=Co-NP. This would be a surprising result since it would prove that the graph isomorphism problem is in NP. Similarly, the discrete logarithm problem is also not known to be NP-complete. Therefore, public key cryptosystems should be fine. Symmetric key cryptosystems such as 3-DES and AES should also be fine. They aren't known to be NP complete problems, and in fact there are no theoretical guarantees about their security at all. They just seem to create messages that are difficult to decrypt given our current cryptoanalytic abilities. In fact, there are NO known NP-complete crypto schemes around today. TL;DR - No encryption schemes will be broken if it is proved that P=NP

is currently in frightened mode

IANAL, but it seems to me that when you buy hardware, you own it. You shouldn't have to agree to an EULA which protects IP because hardware is hardware, not IP. Also, you purchased the computer without reading the EULA, and companies are not allowed to negotiate terms of sale after the fact. Therefore, I don't think this holds up, if I purchased hardware and saw an EULA on it after the fact I would be pissed.