Is a new social media violence monitoring program tracking you?
Kyle S. Reyes | November 8, 2016
(AP Photo/Evan Vucci, File)
It’s almost as if they are worried about a civil war.
Earlier this year, the US Department of Justice (an ironic name, given the department trying to block investigations by the FBI into Hillary Clinton), funded a curious research project.
They allocated $800,000 in funding to a computer program being developed at Cardiff University to scan social media and predict outbreaks in hate crimes.
The announcement came on the same day that Charlotte, North Carolina declared a state of emergency during violent protests after a man was shot dead by police.
The program is meant to find a correlation between what they call “cyber-hate” on Twitter in certain parts of the country and a corresponding offline hate crime. It’s said to help create a predictive model to identify areas where actual crimes are likely to take place after things like the Charlotte shooting.
The program, profiled in the Financial Times, isn’t being built from an artificial intelligence perspective with ANY implicit bias at all.
No, none (sarcasm).
Check out this quote in the article by one of the computer scientists co-leading the project:
“Say a black person was killed by police in the US — and that happens a lot more than it should — we will see biased Tweets coming out, using phrases like ‘They had it coming’ or ‘Get them out’.”
Ok, let’s pause for a second and establish a baseline before we continue.
1) Perpetrators of true hate crimes should be arrested and prosecuted to the fullest extent of the law.
2) Of course there are bad apples in police departments – and every other industry in the world. But comments like “that happens a lot more than it should” are incendiary and unsupported by facts, context or legitimate data.
Where so many artificial intelligence programs are already struggling is in context.
If someone pulls a gun on a cop and is shot and killed…and I say “they had it coming”…does that mean I’m about to commit a hate crime? Does that make me racist? Does that mean I need to worry about getting a knock on the door?
And do we really need an $800,000 investment by the DOJ to tell us that people are going to riot when given a “reason” to loot and destroy their communities and not suffer consequences for it, as we saw in Baltimore, Maryland?
As long as we’re focused more on “words” and not “actions”…we’re simply setting up a more divisive country that creates more problems as opposed to truly finding solutions.
