1) "Stealth" code changes, that could get slipped in via an update, to modify network operation in some unforeseen, but yet-to-conceive exploitable way.
There are more than one Bitcoin Implementations or stacks which work with the Bitcoin Blockchain. Any bugs, backdoors, or problems with one and we can just use the other implementation like libbitcoin.
https://wiki.unsystem.net/en/index.php/Main_Page2) Someone like the World Bank establishing their own, incompatible network.
...coupled with an international media- and marketing campaign, backed by much more finance than the Bitcoin network, perhaps even offering lucrative-by-comparison shares in their new venture, to current Bitcoin players.
Countries have already and are already doing exactly this. Canada, Ecuador are two examples. You shouldn't be concerned about this because :
1) Their digital currencies will likely have security flaws - counterpartry risk from regulators or banks and/or inflation that allows them to steal from the public, and/or doesn't respect users privacy.
2) In the odd event they do create a cryptocurrency that both respects the privacy and property of users than great, we all win anyways, but Bitcoins first mover advantage will probably keep it ahead.
3) Computer intelligence optimizing ("cracking") the protocol
This is unlikely to happen but if this "black swan" event ever did occur than Bitcoin would be the least of anyone's problems as all fiat currency, corporate secrets, and state secrets will be open for everyone. If this ever did happen we could simply take a snapshot of the blockchain and switch algorithms.
4) Isn't it perhaps most relevant how each GH is powered, and how easily that power can be removed by tactical means
5) Add to that security. Who holds this metric: How secure is each GH?
A 51% attack only means that the attacker can do the following :
1) Temporarily prevent a transaction from occurring
2) Create 2-3 false transactions
This attack would quickly get noticed,prevented, and possibly rolled back by the community.