It's a bit orthogonal. User grey-area's main point was that the humans at OpenAI should bear more responsibility for this, not that swarms of semi-intelligent agents are the big danger to be concerned about.