Event Reporting AI?

Would it be possible to develop an AI that keeps record of all news (tv, paper, web) and current events in order to accurately depict what the Jews are doing and why? I think we have the technology available, but nobody has developed it yet. We could redpill the entire world very easily with this kind of development.

Other urls found in this thread:

pastebin.com/raw/UWffBQbn:
github.com/tensorflow/tensorflow/tree/master/tensorflow/core
twitter.com/NSFWRedditGif

Kikes already have that to track bad goy activity. How do you think we get swarmed by bots and shills every time something sensitive is posted?

There would be no way to convey implicit meanings as the computer would only be only able to get things stated explicitly. And that is assuming all outlets tell the truth as they see it, not blatantly lying. For example if one article talks about how Assad bombing all these bakeries is bad, while another talks about how it is a great thing he is bombing them, then the conclusion it will meet is that Assad actually did those things. So all it would take is one outlet being controlled opposition or taking the bait by assuming everyone else is telling the truth to make the AI think it found objective truth.

That is something else though. Usually those are simply trained shills using scripts and bots. What OP is proposing is a program that would read articles from all possible outlets, interpret them and provide detailed information about events that the MSM is trying to bury in disinfo.

AI is not like you think it is. We can barely replicate the brain of a C Elegans.

No, kikes use a wide array of "network analytics companies" to track bad goys online. They also track content that needs to be attacked.

But we could make something similar, in theory, if we had the infrastructure.

It would be like the network spiders that say google uses to compile their search engine. (((The JIDF))) identify problematic websites and have programs like that continually crawl the site looking for keywords so that (((they))) know when to shill (which is infinitely more efficient that doing crawls of the whole internet or manually browsing the board) The military uses it to track that one list of words.

A potential counter is to slightly alter terms the community uses on occasion. I bet the echoes fucked with them hard because they had to rewrite their spiders to include interpretations of the echoes.

So in theory the way we would do it is start compiling a botnet, then setting up on this botnet a network of spiders that crawl various news outlets for a keyword list, and report back to an operator which can then interpret the information (needs a human operator otherwise we don't have the resources to do it) The kikes actually report a lot of truths and alter/remove the information later to suit their purposes.

off topic but this reminds me of a thread on /x/ where someone discovered an AI that did basically this, and I remember it having something to do with the postal service (like if you had been digging to deep the AI would know and you couldn't get a government job)

Or a kind of learning program based on responses. It could be a website with headlines + link to articles. But unlike Reddit and established news websites, no comments, just the thumbs up or down.
The software could then parse out the author,keywords and other meta data from articles, and over time it would be conditioned to show whatever a group thinks is relevant/newsworthy. It could then have links to archived version of said articles. The program itself isn't internally creating context but the users would be in a sense.

Yes. But can you do it in c++?

You mean this one?
pastebin.com/raw/UWffBQbn:

>Implying any other language even has a shot
Sure ofc. If it has a stringent runtime performance requirement (as a big AI system would) most new big systems are built using C++ these days.

any books on this specifically? I'm doing a school project for /po/

what

Surely an AI that is truly sentient could differentiate?

= = = = = = = = = = = = = = = =
Closed Captioning
= = = = = = = = = = = = = = = =
You could do it with broadcast TV news if somehow you could develop an API for closed captioning through cable.
Might not be that hard because closed captioning is drawn after the image on the TV. Its a local thing, so the CC text of the transmission has to be coming through your cable hookup somehow.

= = = = = = = = = = = = = = = =
Ideas
= = = = = = = = = = = = = = = =
You would first have to develop a neural net, and then red-pill that neural network.

You would need to develop a way for the AI to determine if the source of the information was Semitic in nature. This would be easier with print articles due to authors being attributed.

Once you know that the source is Semitic, you would want to put it through a neural network to determine if the article was positive or negative, determine the topic, etc… other qualities about the article.

Creating an AI that can sniff bullshit is bordering on sentience. Not sure if this is possible without quantum computing. We have pretty much peaked as far as current technology will let us create faster and faster computers.

We are making chips so small that they almost short out because there are not enough atoms of insulation between the atoms that conduct.

Would not be hard to track coincidences, like a bunch of newspapers mentioning the same person or event or talking point at the same time. It's a basic analysis but might make a good case that media is essentially controlled by a few people and all says the same thing.

There was a guy on 4cuck that claimed the deep state has super advanced AI that can predict world event with impressive accuracy. Trust me they have that shit.

Bump

It's the 21st century friendo. I can give you the actual software itself:
github.com/tensorflow/tensorflow/tree/master/tensorflow/core

Great thing about TensorFlow (apart from being the world's single most important Deep Learning software) is that you don't have to master Sepples to use it, you can just use Python to drive the actual core software itself.

Want to know more?

How to apply that? Seems like a basic gpu library.
That guy says "kinda" a lot.

Git gud. Learn more about Deep Learning. Learn how to program in either Python, C++, or both (preferably). Finishing that course' videos and going to tensorflow.org and digging through all the examples would be where I would and have begin.

Just be glad you don't have to write the GPU code you are self user.

Yea I know. He's an effete millenial-ish sodomite professor at one of the more liberal (but well known as a successful academic incubator for many high tech startups) trying to communicate to a room full of ultra-nerds and also a few women. What else would he do?

The problem is, it'd take an AI to simply parse all that data from different sources and compile it in a format that can be exploited by such an AI. It's quite a huge task.

And then that shit is CPU hungry. We don't have enough CPU.

Clearly the day we have actually decent AIs, we'll find countless uses to those and they'll solve all our problems. Though when that day comes, everyone will have them and using them will be harder than it'd be today.