Any piece of code will have some degree of vulnerability, and any such code that attracts any level if attention is bound to be exploited through these vulnerabilities.
In conventional (centralised) systems, these vulnerabilities are handled through regular incremental updates that the originator distributes to patch against any vulnerabilities discovered over time.
How are smart contracts vulnerability exploits meant to be secured against over time? Assuming that the code is free from exploits on deployment surely cannot be a solution.