Eric makes a good point in the comments:
The problem with meddlesome preferences on seasteads would be that Iran can always afford to spend more on punishing a Theo Van Gogh than a seastead can spend on protecting him.
I think if we consider the long-term dynamics of a world with both seasteads and land-based states, the problem may be more general and more serious.
If folks are able to choose between living on a seastead or on land, I can imagine an adverse selection-type dynamic causing currently liberalish countries becoming something like big illiberal sects, willing and able to intervene in the affairs of libertarian seasteaders.
If seasteading proves to be more attractive to the liberal-minded, the liberal-minded will disproportionately leave land-based states and the preferences of the median voter will shift in a more meddlesome direction. This will leave the remaining moderately liberal landlubbers dissatisfied, prompting them to leave, and so on. If seasteading became sufficiently popular, this could continue until the median voter of land-based states is significantly more meddlesome than is the case today.
Governments already have the institutions supporting coercive taxation, regulation and conscription, and will thus have higher cooperative efficacy – ability to overcome collective action problems in the production of public goods such as national defence and (widely-desired) aggression. Governments would presumably retain the military resources they have today, will have the ability extract resources from citizens to produce more, and will have the desire to use them.
Liberal seasteads will have trouble defending against aggression by foreign governments, since they will generally be smaller and will be less capable of collectively producing defence against aggression.