The District Attorney in San Francisco will use Artificial Intelligence to make color-blind justice by redacting police reports to delete anything that indicates the race of the accused, so prosecutors will decide which cases to pursue without taint of inherent, even unintentional, racial bias. Is this A.I. decision tool, finally, the fulfillment of the Progressive dream of racial justice?
Categories
Prosecutors Use Artificial Intelligence to Make Color-Blind Justice
The District Attorney in San Francisco will use Artificial Intelligence to make color-blind justice by redacting police reports to delete anything that indicates the race of the accused, so prosecutors will decide which cases to pursue without taint of inherent, even unintentional, racial bias. Is this A.I. decision tool, finally, the fulfillment of the Progressive dream of racial justice?
11 replies on “Prosecutors Use Artificial Intelligence to Make Color-Blind Justice”
We need a new shirt: ” F@ck You, I say What I Want”
Got the idea from Jefferson Airplane, how is that for irony.
One thing we all have to keep in mind is this AI for justice thing is a creature of the Liberals and whatever reason given for using it is almost certainly a lie.
I may need to pause my binge-watching BW. I’ve started to say, “rah-ther” when I debate. ~wink~
Bill, critical thinking was a product of ancient Greek philosophers, and was the foundation of Western civilization. Relative Thinking is the product of the Frankfurt School’s, Critical Theory. It’s an easy mistake, and perhaps an example of the Left co-opting language.
A brief history of critical thinking…
http://www.criticalthinking.org/pages/a-brief-history-of-the-idea-of-critical-thinking/408
I was going to make the same point, but you made it better because you know what you’re talking about, whereas I was going to ask a question. But if I’d talked as much as Bill has on camera, no doubt I would make lots more mistakes than he has ever done!
At 7:00, Scott said, “Isn’t it possible that deep down inside of us we are making decisions in the moment based on biases and bigotry?” The reason non black people look suspiciously at “African Americans” walking toward them in a dark alley is for the very reason Scott cited earlier! Scott said, “African Americans in this 2017 study by the San Francisco District Attorney: ‘African Americans represent only six percent of that county’s population, but accounted for forty-one percent of arrests between … 2008 and 2014.” How can a rational person look at this evidence and then pretend it doesn’t exist?! It’s like the judge admonishing a jury to disregard the testimony they just heard. Absolute poppycock!
There are two kinds of people: those who believe there are two kinds of people and those who don’t. There is another way to divide people: natural and artificial. Natural persons are created by God. Artificial persons are created by natural persons. Artificial persons were first invented by the Romans and then Rome fell. Anything created will ALWAYS be made in the image of its creator. The only way for natural persons to create a sentient being is by natural means. Trying to imitate the brain using neural networks and any other electromechanical means will forever be in the realm of science fiction, no matter how much lip stick you put on this pig.
The good thing about computers though is that they follow a set of deterministic rules, just like the Ten Commandments dictates. But with natural persons, some people are more equal than others. AI can only be unbiased and “color blind” as the natural person who programs it.
So the president needs to take the census argument a step further. Not only should the citizenship question remain, but we should eliminate the boxes for sex and race. After all, from a constitutional standpoint, we only care about the number of citizens, not their sex or ethnic backgrounds.
Yes. You can view things by how they are different, or by how they are similar. In the traditional conservative view, things are considered as to how they are similar. In the progressive Left view, it is always focused on their differences. They claim to be for desegregation, but do everything they can to create the very segregation they claim to hate. Identify politics is the perfect example. The best way to deal with racial bias is, as Morgan Freeman pointed out long ago, STOP TALKING ABOUT IT. Talking about race, refocuses the lens on race. We need to focus it on American citizens, not on race. The only place race should matter anymore is for either medical issues (where race medically does matter) and when it is one way of identifying individuals who have committed a crime (a witness saying the perp is white or yellow or black is valid in helping ID the individual). In terms of victim-hood, and groups, color and race needs to be made verboten. That means no Black Lives Matter groups, no BET network, as they all focus on race. It should be Lives Matter and the Entertainment network. If they want it to focus on a culture, fine, then make it about the culture, i.e., the Hood Entertainment Network, or the Soul Network, or the Backswoods Hillybilly Network, etc. Take the focus off race and put it on culture where it belongs. Racial hatred is what the Left wants because it divides the American people into little special-interest groups they can put at odds with each other and thereby gain power over them all.
That said, I think AI could be used to check for bias of any sort based on the legal facts of the case. But the AIs that we have–or are likely to have have for the foreseeable future–do not have the humanity or compassion to make a completely just decision. They can assist, but they should not be the final arbitration. That must remain a jury of one’s peers, and as good as AIs are, they are not peers, at least not yet. For now, they are simply tools. But they can be great “filters” for bias.
AI = Artificial intelligence = *fake intelligence* = Not real = *fake news*………. Used my AI to deduce this linear path of logic.
They are assuming that racial bias is a factor in prosecutions, but this could backfire in a couple of ways. If bias isn’t a big factor then rates of prosecution of blacks won’t change much. But who is to say there isn’t more racial bias in favor of the Blacks?
I’d like to see their faces if their color blind AI results in even higher prosecution rates of Blacks!
Exactly. I am skeptical that they are actually seeking to remove racial bias. I think they are seeking AI justification for their goal of percentage-based outcomes (i.e., 6% of the population should get only 6% of the prosecutions).
I suspect that, despite how the software is biased (because it is, ipso facto), they will see an increase in the overall number of prosecutions, not just of black folks, and also an increase in the percentage of black folks prosecuted. Because if you really do take out things like socioeconomic advantages (which removing neighborhoods and addresses ought to do) as a proxy for race, you will find that cultures which have a lower police presence (for whatever reason) have higher crime rates. But as with all software and statistics, garbage in, garbage out. They will massage the data to make it the outcomes fit their preconceptions (i.e., their own racial bias), or they will quietly sweep the whole project under the rug.
Should I be wrong and they stand up and shout: “guess what! There is no racial bias, except perhaps a small but statistically significant bias in favor of blacks,” I will gladly, nay joyfully, acknowledge my error.