Probably something like flame does need much money for development. The clearest sign is of course the MD5 collision. For this you need not only the genius who

Probably something like flame does need much money for development. The clearest sign is of course the MD5 collision. For this you need not only the genius who actually devises the attack, but also an entire infrastructure of people who check proofs and chat during coffee breaks on a useful level. So essentially an secret institute for applied cryptography. Atop of this you need to write the rest of the code, which is not at the same level, but still you need at least several programmers. So probably not all suspects are nation-states, but I believe all suspects are rather large organizations.