Even there the standardization process worked: The flaw in Dual EC DRBG was pointed out very quickly, precisely because there was so much cryptanalytic attention paid to it. The fact that RSA (and only RSA, AFAICT) didn't pay attention to the crypto break in exchange for $10MM doesn't change the fact that the vulnerability had been quickly discovered.
If the algorithm had been left unstandardized and was simply "Foo Corp's Custom Wonderful Bit Generator™" the public may never have known of the vulnerability, while the NSA would still have the resources to have discovered the flaw on their own and use TAO to recover the priv key.
ahem... http://en.wikipedia.org/wiki/Dual_EC_DRBG
when you can no longer trust the organizations setting the standards (NIST, CFRG) this argument looses water.