There were a few tells one could find, at least after the fact. For example, he described a women locking her door „with three turns of the key“. That’s something that works on German locks, where the bolt keeps advancing. But the incident was supposed to have happened in some southern US state. And in the US, most locks only allow a quarter turn. I asked specifically about the prevalence of different lock types in the state in a HN threat at the time.

That said, this is a singular and decidedly rare event. The insinuation of widespread whole-cloth fabrication is almost certainly wrong, and probably more damaging to democracy than any actual fraud will ever be.

The three turns is only a salient mess-up in light of other, much more damning evidence. On its own, it's credibly explainable as an honest mistake (I was there but who counts these things? I just assumed).

Yeah I don't know how well this tool could remedy that, or whatever else it hopes to solve. I don't see how this is hardly different than just sourcing your own facts with Google.

Digital signatures from sources would be a nice feature, maybe some kind of trust system such that sources could be kept anonymous.