Verified by 
U.S. officials know that Russians sent advertisements via Facebook and other social media sites to millions of Americans in an attempt to persuade their votes leading up to the 2016 Presidential Election. The majority of these ads were positive towards Donald Trump or were very negative towards Hillary Clinton. The majority also contained false information. People were basically lied to via social media by a group of Russians with a very specific goal.
U.S. officials also know that Cambridge Analytica, a political consultancy firm hired by Trump to help him get elected, stole Facebook data from 50 million individuals. The data were then used to create psychological profiles. From these profiles, this firm would target Facebook users with specific political messages aimed at helping Trump get elected. Cambridge Analytica was basically playing propaganda machine with a massive data set and using advanced statistics to target people for Trump's gain. The leader of this firm was recently recorded by the BBC saying his job was basically to spread poison into the internet and watch it spread. He was also quoted as saying in elections the truth doesn't matter. Emotion matters.
Now all of this is undeniably true, yet you would be hard pressed to find any person who thinks they were influenced by these attempts by these agencies to get Trump elected. The popular retort seems to be "yes but they didn't hack the voting booths." But this ignores that these ads likely very much had an influence—an influence most people, research shows, will either be unaware of entirely or deny.
Emily Pronin, a professor of psychology at Princeton University, coined the term "bias blind spot" to describe people's tendency to think that psychological biases apply more to other people than to themselves. A wide range of studies she and others have conducted have found that, indeed, people will tend to think that things like the hindsight bias (the tendency to think you knew it all along more than you did), cognitive dissonance, and the halo effect (the tendency to think if a person has a positive trait in one area, it will spread across areas) do exist. But, they just apply to others, not the self.
One of the biases is what she describes as thinking you "are alone in a crowd of sheep." Basically, her work shows that people tend to think any form of social influence - from friends, to strangers, to commercials and advertisements—are more likely to influence the decision making of other people than it is to influence their own.
I have done several (as of now unpublished) studies dealing with this particular bias and every study has replicated this basic effect. Interestingly in this case, the question that tends to get the strongest self (not me) vs. other (yes for them though) effect typically involves politics. The scenario will be something like "you are about to vote on a key political issue" and I will ask "How much do you think your vote will be determined by where you live and the people around you?" People will show a strong bias towards thinking this applies more to others than it does to themselves.
I even did a study where I taught the bias to students and the effect still persisted, though every student/participant could accurately identify the bias at the end of the study.
Back to the election.
Now, we have a situation where tens of millions of people (at least) were targeted by groups of people using very sophisticated statistical models (at least in Cambridge's case) and huge data sets to target what would best influence people and then sending those adverts specifically to those people. Then, in line with Pronin's work, denying or even being entirely unaware of the influence this has had on them. It's the perfect propaganda strategy in a (presumably) peaceful democratic system.
You get all of the persuading with none of the awareness of the impact you have had.
