Clearview AI being used to identify Capitol rioters • Hacker News Board

Re: Clearview AI being used to identify Capitol rioters

#171

Post by colejohnson66 » Sun, Jan 10, 2021, 7:19 PM UTC

VBprogrammer wrote: post #119

Barrin92 wrote: post #77

Earlier quoted context omitted.

I'm more concerned with the data being in the hand of an unaccountable private company than with the facial recognition itself to be honest. I remember that Clearview AI itself was associated with far-right individuals last year, so having them police a far-right insurrection from their personal black box system is more than suspect. If it happens, it should be done by an institution that is under the supervision of…

Personally, I'm not really comfortable with the existence of a database of people from which to compare faces against. However, I think it's perhaps more worrying that one day this evidence will be presented in court and the jury will trust the accuracy of "AI" over their own judgment. To clarify, the situation I'm thinking of is where a grainy CCTV image is shown to the jury. The AI expert comes in and talks about a…

Juries already have a tendency to believe the police even when presented with massive evidence to the contrary. Throw in some “expert” testimony from an AI company and we’re in for an even worse situation.

Re: Clearview AI being used to identify Capitol rioters

#172

Post by colejohnson66 » Sun, Jan 10, 2021, 7:20 PM UTC

dragonwriter wrote: post #124

JumpCrisscross wrote: post #82

Earlier quoted context omitted.

> Is it that different to wanted posters or publishing cctv of a crime? I think so. That mechanism is judicially auditable. The AI is not. We should not be arresting people based on the output of an unauditable mechanism.

> That mechanism is judicially auditable. The AI is not. We should not be arresting people based on the output of an unauditable mechanism. Are we, though? If “ClearView AI gave us a hit” is being treated as probable cause sufficient for an arrest warrant, that's a problem. If it's being used as a tool to generate leads to investigate and traditional evidence is gathered and presented, I don't see a big problem.

Police have demonstrated multiple times that “a hit” is enough for them to railroad an innocent person

Re: Clearview AI being used to identify Capitol rioters

#173

Post by wpietri » Sun, Jan 10, 2021, 10:02 PM UTC

wpietri wrote: post #142

mmmBacon wrote: post #17

Earlier quoted context omitted.

I think we already have a system in the West where people are afraid to speak openly; it’s referred to as cancel culture but exists elsewhere. The difference is that it’s not a formal government system but more of a mob social justice system.

Oh, please. People have been "afraid to speak openly" for all of America's history. The notion that there's something drastically different about "cancel culture" in terms of severity is historically ignorant. Look at the comedian Lenny Bruce's arrests for obscenity, for example. Or the anticommunist panic of the 1950s. To say nothing of how women were controlled up to and after suffrage: deprivation of rights, margi…

I think it is hilarious when, as happens all the time here on HN, people against "cancel culture" downvote critics in hopes of canceling speech they don't like.

Re: Clearview AI being used to identify Capitol rioters

#174

Post by j8hn » Sun, Jan 10, 2021, 10:10 PM UTC

heyoni wrote: post #168

j8hn wrote: post #121

Earlier quoted context omitted.

"The war on the female reproductive system". I realize you are referring to pro-choice. I'm curious why you chose to phrase it in this way. What is the war on education?

I'm not sure why I chose to phrase it that way. Probably because it's ongoing, multifaceted and victim to some very very dirty tactics. Also, we like "waging war" on things here rather than solve them. Briefly, our education system in the US is one of the least well performing systems amongst modern states. It is also the least well funded systems with conservatives trying to practically hand it over to private schoo…

Is the GOP platform to privatize schools? If it was, how is that a war on education?

Re: Clearview AI being used to identify Capitol rioters

#175

Post by jackson1442 » Sun, Jan 10, 2021, 10:46 PM UTC

inglor_cz wrote: post #63

jackson1442 wrote: post #35

Earlier quoted context omitted.

In reality, cancel culture is just the free market at work and has always felt, to me, like a narrative peddled by the powerful to allow themselves to not be held accountable for their actions. If I don't want to buy a Musk-mobile because I don't agree with Musk on issues like racism[0] or don't like how he treats his employees[1], I am allowed to act on it. I am also allowed to share these thoughts with people on ap…

"If you want to post racist/sexist/homophobic things on your public Twitter, so be it, but don't be surprised when someone finds it." So, what is going to be the kiss of death in 2050? Do you know in advance? How many of your contemporary comments are going to run afoul of the standard of 2050 and will you remember to delete them all, including from Internet archive/wayback sites, in 2049? Remember, a senior manager…

I definitely agree that his forced resignation wasn't exactly just, but that's not really cancel culture in action, that was a response to a single employee complaint from my research. It almost reads as if execs were looking for an excuse to get rid of him. If it was up to me, I'd just quietly release a statement or something along that... 33 years is a long time.

What I'm more referring to is situations like JK Rowling where she is openly posting hateful content publicly (people don't even have to dig for it!) and she's criticizing cancel culture as the reason people don't want to read her new book, which just looks like it'll be awful[1].

[1] https://www.cnn.com/2020/09/15/entertainment/jk-rowling-trou...

Re: Clearview AI being used to identify Capitol rioters

#176

Post by jtsiskin » Mon, Jan 11, 2021, 4:27 AM UTC

renewiltord wrote: post #148

jtsiskin wrote: post #80

Earlier quoted context omitted.

I really can’t tell if this is sarcasm, satire, or serious. I think satire?

I get this a lot with this viewpoint but I've passed through the other ones to get to this one. If it helps, I am familiar with the Veil of Ignorance, the Categorical Imperative, and every other basic tool of ethics you can think of.

Do you think you would still have the same opinion if you were black? Even if selective enforcement makes sense, I would want very smart, highly educated, trusted, and fair people, who know everything about the defendant, making the decision. Not a cop, with nothing more than a photo, choosing who to pursue

Re: Clearview AI being used to identify Capitol rioters

#177

Post by renewiltord » Mon, Jan 11, 2021, 4:45 AM UTC

jtsiskin wrote: post #176

renewiltord wrote: post #148

Earlier quoted context omitted.

I get this a lot with this viewpoint but I've passed through the other ones to get to this one. If it helps, I am familiar with the Veil of Ignorance, the Categorical Imperative, and every other basic tool of ethics you can think of.

Do you think you would still have the same opinion if you were black? Even if selective enforcement makes sense, I would want very smart, highly educated, trusted, and fair people, who know everything about the defendant, making the decision. Not a cop, with nothing more than a photo, choosing who to pursue

Probably not, but like I said, I'm not, in fact, black. Like I said, I'm familiar with the Veil of Ignorance, so you can assume I also know affiliated concepts.

Re: Clearview AI being used to identify Capitol rioters

#178

Post by scsilver » Mon, Jan 11, 2021, 4:11 PM UTC

ianai wrote: post #108

zozbot234 wrote: post #104

Earlier quoted context omitted.

The sad thing is precisely that no-one cares about "re-integrating" proponents of right-leaning views "to society". Everyone is talking about forcibly deplatforming them all out of society which, even if successful, will only make them far more marginalized and far more open to dangerous, violent extremism. Is this really what we need? We've had enough divisiveness already, so why not try a more constructive approach…

That is decidedly opposite of how I see the immediate political inertia. It might help to revisit previous times the democrats took over all three legislative branches. They’re not unified ever and dissent is assured. Meaning there are conservative wings which will welcome and work with the far right even after this. If you’re referencing the deplatforming going on from corporate entities that’s a whole another pheno…

They can start by not spreading covid to our grandparents.

Re: Clearview AI being used to identify Capitol rioters

#179

Post by pvaldes » Mon, Jan 11, 2021, 11:25 PM UTC

walls wrote: post #54

jonas21 wrote: post #37

Earlier quoted context omitted.

Out of curiosity, why do you feel like you're supposed to be supportive of the use of facial recognition in this case?

Identifying terrorists is a positive outcome.

Identifying cover agents infiltrated into the mob, for example would be a not so positive one. Once this cat is out of the box, is out. Can't be entered again. All that you need is money to buy the product, and narcos have a lot of it.