{"id":14520,"date":"2019-03-21T23:46:03","date_gmt":"2019-03-21T23:46:03","guid":{"rendered":"https:\/\/rsoftsolution.com\/blog\/facebooks-ai-couldnt-spot-mass-murder.html"},"modified":"2019-03-21T23:46:03","modified_gmt":"2019-03-21T23:46:03","slug":"facebooks-ai-couldnt-spot-mass-murder","status":"publish","type":"post","link":"https:\/\/www.rbsitsoftwaresolution.in\/blog\/facebooks-ai-couldnt-spot-mass-murder.html","title":{"rendered":"Facebook\u2019s AI couldn\u2019t spot mass murder"},"content":{"rendered":"

Facebook has given another update<\/a> on measures it took and what more it\u2019s doing in the wake of the livestreamed video of a gun massacre<\/a> by a far right terrorist who killed 50 people in two mosques in Christchurch, New Zealand.<\/p>\n

Earlier this week<\/a> the company said the video of the slayings had been viewed less than 200 times during the livestream broadcast itself, and about about 4,000 times before it was removed from Facebook \u2014 with the stream not reported to Facebook until 12 minutes after it had ended.<\/p>\n

None of the users who watched the killings unfold on the company\u2019s platform in real-time apparently reported the stream to the company, according to the company.<\/p>\n

It also previously <\/a>said it removed 1.5 million versions of the video from its site in the first 24 hours after the livestream, with 1.2M of those caught at the point of upload \u2014 meaning it failed to stop 300,000 uploads at that point.\u00a0Though as we pointed out in our earlier report<\/a> those stats are cherrypicked \u2014 and only represent the videos Facebook identified. We found other versions of the video still circulating on its platform 12 hours later.<\/p>\n

In the wake of the livestreamed terror attack, Facebook has continued to face calls from world leaders to do more to make sure such content cannot be distributed by its platform.<\/p>\n

The prime minister of New Zealand, Jacinda Ardern told media yesterday<\/a> that the video \u201cshould not be distributed, available, able to be viewed\u201d, dubbing it: \u201cHorrendous.\u201d<\/p>\n

She confirmed Facebook had been in contact with her government but emphasized that in her view the company has not done enough.<\/p>\n

She also later told the New Zealand parliament: \u201cWe cannot simply sit back and accept that these platforms just exist and that what is said on them is not the responsibility of the place where they are published. They are the publisher. Not just the postman.\u201d<\/p>\n

We asked Facebook for a response to Ardern\u2019s call for online content platforms to accept publisher-level responsibility for the content they distribute. Its spokesman avoided the question \u2014 pointing instead to its latest piece of crisis PR<\/a> which it titles: \u201cA Further Update on New Zealand Terrorist Attack\u201d.<\/p>\n

Here it writes that \u201cpeople are looking to understand how online platforms such as Facebook were used to circulate horrific videos of the terrorist attack\u201d, saying it therefore \u201cwanted to provide additional information from our review into how our products were used and how we can improve going forward\u201d, before going on to reiterate many of the details it has previously put out.<\/p>\n

Including that the massacre video was quickly shared to the 8chan message board by a user posting a link to a copy of the video on a file-sharing site. This was prior to Facebook itself being alerted to the video being broadcast on its platform.<\/p>\n

It goes on to imply 8chan was a hub for broader sharing of the video \u2014 claiming that: \u201cForensic identifiers on many of the videos later circulated, such as a bookmarks toolbar visible in a screen recording, match the content posted to 8chan.\u201d<\/p>\n

So it\u2019s clearly trying to make sure it\u2019s not singled out by political leaders seek policy responses to the challenge posed by online hate and terrorist content.<\/p>\n

Further details it chooses to dwell on in the update is how the AIs it uses to aid the human content review process of flagged Facebook Live streams are in fact tuned to \u201cdetect and prioritize videos that are likely to contain suicidal or harmful acts\u201d \u2014 with the AI pushing such videos to the top of human moderators\u2019 content heaps, above all the other stuff they also need to look at.<\/p>\n

Clearly \u201charmful acts\u201d were involved in the New Zealand terrorist attack. Yet Facebook\u2019s AI was unable to detected a massacre unfolding in real time. A mass killing involving an automatic weapon slipped right under the robot\u2019s radar.<\/p>\n

Facebook explains this by saying it\u2019s because it does not have the training data to create an algorithm that understands it\u2019s looking at mass murder unfolding in real time.<\/p>\n

It also implies the task of training an AI to catch such a horrific scenario is exacerbated by the proliferation of videos of first person shooter videogames on online content platforms.<\/p>\n

It writes: \u201c[T]his particular video did not trigger our automatic detection systems. To achieve that we will need to provide our systems with large volumes of data of this specific kind of content, something which is difficult as these events are thankfully rare. Another challenge is to automatically discern this content from visually similar, innocuous content \u2013 for example if thousands of videos from live-streamed video games are flagged by our systems, our reviewers could miss the important real-world videos where we could alert first responders to get help on the ground.\u201d<\/p>\n

The videogame element is a chilling detail to consider.<\/p>\n

It suggests that a harmful real-life act that mimics a violent video game might just blend into the background, as far as AI moderation systems are concerned; invisible in a sea of innocuous, virtually violent content churned out by gamers. (Which in turn makes you wonder whether the Internet-steeped killer in Christchurch knew \u2014 or suspected \u2014 that filming the attack from a videogame-esque first person shooter perspective might offer a workaround to dupe Facebook\u2019s imperfect AI watchdogs.)<\/p>\n

Facebook post is doubly emphatic that AI is \u201cnot perfect\u201d and is \u201cnever going to be perfect\u201d.<\/p>\n

\u201cPeople will continue to be part of the equation, whether it\u2019s the people on our team who review content, or people who use our services and report content to us,\u201d it writes, reiterating yet again that it has ~30,000 people working in \u201csafety and security\u201d, about half of whom are doing the sweating hideous toil of content review.<\/p>\n

This is, as we\u2019ve said many times before, a fantastically tiny number of human moderators given the vast scale of content continually uploaded to Facebook\u2019s 2.2BN+ user platform.<\/p>\n

Moderating Facebook remains a hopeless task because so few humans are doing it.<\/p>\n

Moreover AI can\u2019t really help. (Later in the blog post Facebook also writes vaguely that there are \u201cmillions\u201d of livestreams broadcast on its platform every day, saying that\u2019s why adding a short broadcast delay \u2014 such as TV stations do \u2014 wouldn\u2019t at all help catch inappropriate real-time content.)<\/p>\n

At the same time Facebook\u2019s update makes it clear how much its \u2018safety and security\u2019 systems rely on unpaid humans too: Aka Facebook users taking the time and mind to report harmful content.<\/p>\n

Some might say that\u2019s an excellent argument for a social media tax<\/a>.<\/p>\n

The fact Facebook did not get a single report of the Christchurch massacre livestream while the terrorist attack unfolded meant the content was not prioritized for \u201caccelerated review\u201d by its systems, which it explains prioritize reports attached to videos that are still being streamed \u2014 because \u201cif there is real-world harm we have a better chance to alert first responders and try to get help on the ground\u201d.<\/p>\n

Though it also says it\u00a0expanded its acceleration logic last year to \u201calso cover videos that were very recently live, in the past few hours\u201d.<\/p>\n

But again it did so with a focus on suicide prevention \u2014 meaning the Christchurch video would only have been flagged for acceleration review in the hours after the stream ended if it had been reported as suicide content.<\/p>\n

So the \u2018problem\u2019 is that Facebook\u2019s systems don\u2019t prioritize mass murder.<\/p>\n

\u201cIn [the first] report, and a number of subsequent reports, the video was reported for reasons other than suicide and as such it was handled according to different procedures,\u201d it writes, adding it\u2019s \u201clearning from this\u201d and \u201cre-examining our reporting logic and experiences for both live and recently live videos in order to expand the categories that would get to accelerated review\u201d.<\/p>\n

No shit.<\/p>\n

Facebook also discusses its failure to stop versions of the massacre video from resurfacing on its platform, having been \u2014 as it tells it \u2014 \u201cso effective\u201d at preventing the spread of propaganda from terrorist organizations like ISIS with the use of image and video matching tech.<\/p>\n

It claims\u00a0 its tech was outfoxed in this case by \u201cbad actors\u201d creating many different edited versions of the video to try to thwart filters, as well as by the various ways \u201ca broader set of people distributed the video and unintentionally made it harder to match copies\u201d.<\/p>\n

So, essentially, the \u2018virality\u2019 of the awful event created too many versions of the video for Facebook\u2019s matching tech to cope.<\/p>\n

\u201cSome people may have seen the video on a computer or TV, filmed that with a phone and sent it to a friend. Still others may have watched the video on their computer, recorded their screen and passed that on. Websites and pages, eager to get attention from people seeking out the video, re-cut and re-recorded the video into various formats,\u201d it writes, in what reads like another attempt to spread blame for the amplification role that its 2.2BN+ user platform plays.<\/p>\n

In all Facebook says it found and blocked more than 800 visually-distinct variants of the video that were circulating on its platform.<\/p>\n

It reveals it resorted to using audio matching technology to try to detect videos that had been visually altered but had the same soundtrack. And again claims it\u2019s trying to learn and come up with better techniques for blocking content that\u2019s being re-shared widely by individuals as well as being rebroadcast by mainstream media. So any kind of major news event, basically.<\/p>\n

In a section on next steps Facebook says improving its matching technology to prevent the spread of inappropriate viral videos being spread is its priority.<\/p>\n

But audio matching clearly won\u2019t help if malicious re-sharers just both re-edit the visuals and switch the soundtrack too in future.<\/p>\n

It also concedes it needs to be able to react faster \u201cto this kind of content on a live streamed video\u201d \u2014 though it has no firm fixes to offer there either, saying only that it will explore \u201cwhether and how AI can be used for these cases, and how to get to user reports faster\u201d.<\/p>\n

Another priority it claims among its \u201cnext steps\u201d is fighting \u201chate speech of all kinds on our platform\u201d, saying this includes more than 200 white supremacist organizations globally \u201cwhose content we are removing through proactive detection technology\u201d.<\/p>\n

It\u2019s glossing over plenty of criticism on that front too though \u2014 including research that suggests<\/a> banned far right hate preachers are easily able to evade detection on its platform. Plus its own foot-dragging on shutting down far right extremists. (Facebook only finally banned<\/a> one infamous UK far right activist last month, for example.)<\/p>\n

In its last PR sop, Facebook says it\u2019s committed to expanding its industry collaboration to tackle hate speech via the Global Internet Forum to Counter Terrorism (GIFCT), which formed in 2017<\/a> as platforms were being squeezed by politicians to scrub ISIS content \u2014 in a collective attempt to stave off tighter regulation.<\/p>\n

\u201cWe are experimenting with sharing URLs systematically rather than just content hashes, are working to address the range of terrorists and violent extremists operating online, and intend to refine and improve our ability to collaborate in a crisis,\u201d Facebook writes now, offering more vague experiments as politicians call for content responsibility.<\/p>\n

Article source: http:\/\/feedproxy.google.com\/~r\/techcrunch\/social\/~3\/meSa5tx64uo\/<\/p>","protected":false},"excerpt":{"rendered":"

Facebook has given another update on measures it took and what more it\u2019s doing in the wake of the livestreamed video of a gun massacre by a far right terrorist <\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"jetpack_post_was_ever_published":false,"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2}},"categories":[3],"tags":[10,7,14,15,6,12,11],"class_list":["post-14520","post","type-post","status-publish","format-standard","hentry","category-latest-update","tag-computer-news","tag-downloads","tag-microsoft-download-news","tag-networking-updates","tag-technology-update","tag-wordpress-plugin-download","tag-wordpress-theme-download"],"aioseo_notices":[],"aioseo_head":"\n\t\t\n\t\n\t\n\t\n\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t\n\t\t