2.4. What makes you think you’ll be seen as an objective source?
The roots of this system were developed and proven effective in the financial markets; an
environment that instantly and ruthlessly punishes anything except absolute and total objectivity.
Beyond that, we believe our results will make it very clear. In our testing, we found that we
have people who score both high and low on the political left and the political right.
Even more important, there were people on both ends of the spectrum who surprised us. People
who we thought of as being quite far out on the tails of the political spectrum who scored
comparatively well in our testing, and people we thought of as mostly middle of the road who
In data science, if everything turns out exactly the way you expect, that’s likely to be an indicator
of a problem called confirmation bias. That’s a difficult thing to get a handle on so we worry
But when we went back and looked at the twitter feeds of the people we thought were extreme, it
always turned out that online at least, their posts were moderate and thoughtful, and the inverse
was true for our surprise low scorers. So whatever our initial impressions of the person in
question, the system was reading what they said and assessing it appropriately.
There’s something else too. If all we did was read the news and declare a story fake or real, how
would we know? The only thing we’d have to compare it to is our own viewpoint and that
viewpoint contains bias. This is the same mistake all the other ‘fake news’ solutions are making.
Some, thought clearly not all, are just political opposition and are doing that very thing on
purpose to promote political propaganda.
The only thing we compare a score to is everyone else’s score. In that regard our system is being
‘objective’ in the only way anything ever can be. It isn’t a reflection of our views, it’s an
accurate reflection of everyone’s views. You might not agree with that, but agree or not, it’s