We as humans have a lot of intuitive tools for knowing whether a source of data is trustworthy. AI could possibly approach this ability given enough training...

We as humans have a lot of intuitive tools for knowing whether a source of data is trustworthy. AI could possibly approach this ability given enough training... we'd need to do something like add a "trust" score to every node in the graph.