Beginning stages of ai scanning photos

By showing it millions of photos and having us show it how to locate a key term, it can begin to learn what the key terms attributes are

The end goal here is meme tracking. It was specifically mentioned in the conclusion of that "Jow Forums meme study" a couple years ago. They can currently track and index word based memes ("praise kek") by scraping archives and image based memes (pepe/wojak) via image hashing but they've been helpless to so much as organize screencaps and original content. This is their attempt at figuring out how to monitor and contain them.

Attached: 1558819054668.png (1024x1024, 923K)

I feel like they wont be able to keep up with the evolution of humor

Well it looks like these fucking plebbitors better start contributing some OC and leaving the boomer memes on faceberg.

I somewhat see why you have come to this conclusion. But we havent made the leap to helping ai identify human nature (at least thru captcha) I think we've been training ai thru the simple act of shitposting

m.youtube.com/watch?v=snig1RUtZos

The bad guys will always be behind.

Attached: checkem.png (800x450, 606K)

Maybe they are teaching their robot army to read street signs, secure streets and fire hydrants, disable cars and bikes

>Limitations & Future Work:
>Overall, our analysis not only pro- vides the first measurement study of Jow Forums, but also insight on the continued growth of hate and extremism trends in online social me- dia, and highlights a few interesting problems for future research. Naturally, our work is not without limitations, and a few caveats should be considered when looking at our findings. First, although the Hatebase dataset is an invaluable resource for our analysis, it is merely a dictionary while the usage of these words is often context- dependent. We leave it to future work to investigate how to distin- guish context (e.g., by recognizing sarcasm or trolling). Second, our flag and country analysis may have been influenced by the use of VPNs/proxies: although this does not affect the validity of our results, it calls for a more in-depth analysis of language and posting behavior. Finally, while we showed quantitative evidence that raids are taking place, we make no claims about our ability to classify them as there are many layers of subtlety in how raiding behavior might be exhibited, and our findings really serve as a springboard for future work, rather than a readily available classifier.
semanticscholar.org/paper/A-Longitudinal-Measurement-Study-of-Jow Forums's-Forum-Hine-Onaolapo/745c5699872ad7e469758224d240d0afe8b49dfc

When are we going to adress the fact that AI is the inevitible doom of us?

The end of free will, it can already read our minds

If its reading our minds cant we influence it rather than the other way around?

this

omg literally so much THIS

I do this too. I dont know how much effect it will have, but its fun to think is subverting a mega corporation's ai project.

Attached: cheeky frog poorly drawn.png (396x400, 26K)

I was looking around the house looking for something to open a bottle of wine, I didnt say anything about it out loud, so after 20 minutes I go to youtube and type "how to" and it autofilled open a wine bottle as the first auto fill, the even more fucked up thing is I didnt purchase it, it was a gift, I could see if my rfid chip showed I purchased wine that that autofill would make sense, but these fucking things can read minds

Facebook has been doing this for years. Not news. You can't stop the electronic jew. The only solution is to grill, drink your own beer, listen to classic rock, and mow the lawn. Jews hate that.

Hard to say, but as long as it's accepting the wrong input, then I assume it's working.

Dont forget, fucking of age women. Thats something they don't like taking part in

I keep tabs on this shit just as a precaution because its better to be aware than blind but honestly Im not too concerned that theyll achieve their Orwellian end game of stopping "hate speech" upon its creation in various forms online. Im just thinking back to when the curtain first got pulled back on old /b/ when it was for the first time not so much a secret club and being lurked and white knighted by journalist/SJWs

Just because they find the room or even made a key to the door we've been metaphorically raping their ideologies in that doesn't mean they can prevent anons from continuing to do so. All they can do is sit back and watch like the little faggot cucks they are.

Attached: 1554391368833.jpg (640x427, 44K)

Attached: 1530247826649.png (409x284, 105K)

No but im sure they will try to collapse our society because their feelings are hurt

It's bullshit, the items are already identified in captchas. It already knows which square is a fire hydrant and which is a taxi. This is how it knows if you're right or not.

Youre retarded, you can pick incorrect squares and still be allowed to post

They only win if you stop shit posting. That is the achilles heel in their Master plan they don't seem to understand. They could spend a trillion dollars on drones with cameras and microphones to follow shit posters around all day every day, but no technology will ever be capable of stopping people from flipping them off and calling it a robo nigger and making people mock them for it.

People will always find a workaround to say banned words.

Pretty sure I've seen someone named NillKiggers on TF2.

They will never detect my abstract merchants

Attached: e11.png (171x184, 5K)

True, self censorship is the end goal, only problem is the normies are so fucking retarded they will excommunicate

Good idea.
As I see it there's too many work arounds.
Soon enough theyll realize it.

Too much effort has been spent by them.
However you look at it they're doomed.
Everyone can see it but them apparently.

Kids will always invent new methods.
Innovation comes naturally.
Killing free speech is impossible.
Evidence is everywhere.
Stupid bastards.

Yes but they have been pretty successful brainwashing the retards to do their censoring for them

It takes a small army of moderators to scrub wrong think from an online forum. It only takes 1 autist with a catchy meme to put them back to square 1.

Im talking about the retards on social media

i always just mark the 3 stock pictures and the one that the ai wants to find i never mark. so you pass the captcha test but the ai geta no information to learn

maybe your bot scor just goes down because youcan't distinguish between buses and 5rucks while a million normies find them easily. so they have enough data and you just lose google points to the point that you are trapped in captcha loops

IIRC Kikebook outsourced some of their moderation to a company who pays people $29k a year to moderate shit there. And people managed to get PTSD from that.

I don't get it, I can't see the merchant at all.

Attached: 37ab5bc63d98ac529c421fc6913a5765051d556095bf785799715e4984e08d0a.png.cf.png (368x376, 3K)

Because normies don't have the iq to handle the truth so they end up going brain dead and get put on pharmaceuticals. And guess who owns the pharmaceutical companies

Attached: I wonder who the lines represent.png (310x253, 3K)

Rebel

>breaking news update: antisemitic graffiti responsible for riots at Berkeley college over the weekend determined to be coicendentally smeared cigarette ashes...or were they?!

Attached: 20190726_121827.jpg (720x729, 291K)

We can go simpler.

Attached: I wonder who the lines represent 2.png (444x444, 2K)

We can go alot simpler

Attached: 1534702076039.png (461x583, 2K)

Real hate speech right there.

The facebook ai scanning is surprisingly precise. Like, every other fun little site that allows to input photos/videos and uses ai to detect things on them is clunky and doesn't give good answers, but when it's facebook, where the scanning process is happening without your agreement or knowledge, it's super precise.

Attached: 67280640_2295456684049968_3503247043428614144_n.jpg (624x960, 112K)

>without your agreement or knowledge
nigga Facebook has full ownership of your pictures once they're posted, so zuck can literally photoshop a dick in your mouth and post it and there's nothing you can do about it

The example shown is instagram on google chrome

false data is ignored unless millions of people were to coordinate to feed it the exact same false data. Which won't happen.

facebook is doing this for 3-4 years.
today even your mobile camera knows to detect faces and suroundings to adjust settings.
boomer or shill

I don't know. I've had Capcha insist that something that wasn't a crosswalk was a crosswalk.

It keeps introducing new images and making assumptions about them that get corrected by people over time in the cases where it's wrong.

will it finally be use to rid us of the pedo scum?

Dig on photomyne, based in Tel-Aviv.

Fucking captcha