#2 is certainly food for thought. So the idea is that from a journalistic fact-checking point of view, it is more important to convey the information exactly as it was presented than to verify its accuracy?
This would explain why science/engineering-based articles are so commonly inaccurate or missing in critical details. The journalist can fall back on saying "I have a recording of an interview with the expert after we downed a few pints at the pub, and I'm just parroting back what he said. Don't shoot the messenger!"