I think most people understand the value of propaganda. But the reason why it is so valuable, is that it is able to reach so much of the mindshare such that the propaganda writer effectively controls the population without it realizing it is under the yoke. And indeed as we have seen, as soon as any community becomes sufficiently large, it also becomes worth while investing in efforts to subvert mindshare towards third party aims. Both in person and online communities.

AI is no different in this regard. Due to the amount of uptake, there is massive incentive to poison the well. Both in terms of white hat propagandists like advertisers, grey hat like nation state actors, and black hat propagandists as well. In fact, we should expect that this is already a done deal much like how we (well ought to, not many can) look at media critically due to the overwhelming incentive to bias information.

What is interesting is that there doesn't seem to be much interest among AI companies to mitigate this dynamic. Maybe there is no real way that this dynamic can ever be mitigated. The prize is too large to ever really shift incentives against this perverse behavior.

Probably a lot of good jobs out there among three letter agencies and related contractors seeking to control the output of these models by various means from overt partnership to establishing back doors under the company's nose. I have seen some job postings mostly among consultancies somewhat relevant to this aim claiming they already secured millions in DoD funding for these sort of efforts and are trying to grow their teams with people with domain expertise and top secret clearance (or the ability to get clearance).