The thing is, the people involved in these orgs, many will go to their graves thinking they were doing the right thing, believing themselves to be on a mission guided by a secret higher truth - or to protect the common-good - or maintain societal stability.

I agree, and this is an important point. Take this paper for example:

Generative Language Models and Automated Influence Operations: Emerging Threats and Potential Mitigations

It's a fascinating topic and one worthy of research, in fact I'd say understanding how AI can generate propaganda is essential knowledge in these times.

But a central question that the paper doesn't seem to address (I've only skimmed and scanned), is who gets to decide what qualifies as propaganda?

The people doing this research probably think they are saving the world from evil forces, but the science they are doing will certainly not be applied objectively. It will be used to serve somebody's purposes.