If all this data would be more or less random (which is kind of implied by "run on bots worldwide in millions") this would merely add some random noise to large datasets. For this to create any real problems the noise generated by those bots would have to be somehow biased, of course in an incorrect way. This implies then, that for such idea to be even remotely effective one would need at least a peak in aggregated data to know real trends and program bots to blur the image generating fake trends.
Thank you for your reply. Granted that Google, Microsoft, Facebook, and all the others collect data on us. And this is used by the Government security agencies (i.e. NSA/CIA as described by Snowden) to spy on us -- as we all know now the agencies have backdoors to these Corporations enabled with stuff like 'secret rooms' (AT&T) common. In which case this is used to detect terrorism threats. Why state-operators with enormous funding, like let's say Iran or North Korea or Russia or China -- won't generate systems like this to make NSA/CIA job harder (if anything). If I were Iran state security institution planning, targeting and executing terrorist attacks against the US and their allies -- I would be very interested to do things like millions of email accounts sending millions of emails daily with trigger words like 'bomb', 'attack', 'death to america', etc, etc. So my real communication would be a drop in the ocean of similar-looking noise.
What is stopping these actors from doing just that?
I have wondered about this for long time and I'm just curious. If the whole 'terrorism threat' thing is real, we would see counter-actions like this done by terrorism sponsoring organizations or states. No?
