The Trump campaign used Steve Bannon's Cambridge Analytica company to profile and target 3.5 million Black voters in 2016 to stop them from voting for Hillary Clinton, according to UK Channel 4 News.
Channel 4 News has exclusively obtained a vast cache of data used by Donald Trump’s 2016 presidential campaign on almost 200 million American voters.
It reveals that 3.5 million Black Americans were categorised by Donald Trump’s campaign as ‘Deterrence’ – voters they wanted to stay home on election day.
Tonight, civil rights campaigners said the evidence amounted to a new form of voter “suppression” and called on Facebook to disclose ads and targeting information that has never been made public.
The ‘Deterrence’ project can be revealed after Channel 4 News obtained the database used by Trump’s digital campaign team – credited with helping deliver his shock victory to become president four years ago.
Vast in scale, it contains details on almost 200 million Americans, among more than 5,000 files, which together amass almost 5 terabytes of data – making it one of the biggest leaks in history.
It reveals not only the huge amounts of data held on every individual voter, but how that data was used and manipulated by models and algorithms.
In 16 key battleground states, millions of Americans were separated by an algorithm into one of eight categories, also described as ‘audiences’, so they could then be targeted with tailored ads on Facebook and other platforms.
One of the categories was named ‘Deterrence’, which was later described publicly by Trump’s chief data scientist as containing people that the campaign “hope don’t show up to vote”.
Analysis by Channel 4 News shows Black Americans – historically a community targeted with voter suppression tactics – were disproportionately marked ‘Deterrence’ by the 2016 campaign.
In total, 3.5 million Black Americans were marked ‘Deterrence’.
In Georgia, despite Black people constituting 32% of the population, they made up 61% of the ‘Deterrence’ category. In North Carolina, Black people are 22% of the population but were 46% of ‘Deterrence’. In Wisconsin, Black people constitute just 5.4% of the population but made up 17% of ‘Deterrence’.
The disproportionate categorising of Black Americans for ‘Deterrence’ is seen across the US. Overall, people of colour labelled as Black, Hispanic, Asian and ‘Other’ groups made up 54% of the ‘Deterrence’ category. In contrast, other categories of voters the campaign wished to attract were overwhelmingly white.
The 2016 campaign preceded the first fall in Black turnout in 20 years and allowed Donald Trump to take shock victories in key states like Wisconsin and Michigan by wafer-thin margins, reaching the White House despite losing the popular vote to Hillary Clinton.
Trump’s digital campaign, called ‘Project Alamo’ and based in San Antonio, Texas, involved a team from the now defunct British company Cambridge Analytica, working with a team from the Republican National Committee. Two senior members of the Cambridge Analytica team are working on the Trump 2020 campaign.
Cambridge Analytica collapsed after investigations by Channel 4 News, The Observer and the New York Times in 2018.
We knew Cambridge Analytica was doing this back in 2017. The new information is that they were working directly with the Trump campaign and with Facebook to build these voter profiles of 200 million Americans and targeting them with such pinpoint accuracy that Trump won exactly the states he needed by exactly the margins he needed.
And now we know precisely what the purpose of those profiles were: to categorize tens of million of Americans in order to find and attract the voters they needed with Facebook and social media, and suppress those they needed to stay home in key states with targeted ads.
They nailed the entire electorate. They divided us like never before. They used this and voter ID laws and they got away with it. They had our number, all 200 million of us, and bought an election.
Here's the thing though. For this deterrence plan to work, and it did, Black turnout was cut by 19% overall in PA, MI, WI and OH, for this to have worked at all, it had to be close enough for them to cheat.
It never should have been close enough for them to do so.
But Trump voters made it so.