Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Twitter to label deepfakes and other deceptive media

reuters.com

Post reply on HN
Print view View original

91–100 of 151 posts

Re: Twitter to label deepfakes and other deceptive media

#91

Post by rhcom2 »

mc32 wrote: post #45
pjc50 wrote: post #35

There are obvious cases of deceptive media, where things are simply made up, but the most invidious ones are just "differential spin" and very hard to pin down. The recent Harry/Megan fiasco has provided a great set of examples on how the same action can be reported as a positive or negative depending on the paper's preferred spin: https://www.buzzfeednews.com/article/ellievhall/meghan-markl...

I saw a side by side of an opinion editor of the WaPo. Last year she was saying how important the Iowa caucuses were. This year she said how unimportant they are. The same person, diametric takes on the same subject. The only difference is I guess her preferred candidate didn’t do as well...

The charitable interpretation is to include the other difference, which is time. People change their opinions.

Re: Twitter to label deepfakes and other deceptive media

#92

Post by dTal »

bilekas wrote: post #74
dTal wrote: post #60

Earlier quoted context omitted.

This seems to be a language drift worth nipping in the bud. It's not "fake". It's a lie. When something purports to be something it's not, "fake" is an adjective describing that thing. "Lie" is the adjective that describes the claim . If you tweet that you are Man of the Year 2020, you are fake, a phony. The tweet , on the other hand, is simply a lie.

But the content is the article in this case. The content is fake.. No ? Maybe I have messed up the semantics. I'm not trying to be difficult, but is it not valid in this context to correlate "lie" with "fake" ?

There is overlap, but the distinction is that "fake" things contain implicit lies about their nature.

Your example, a correctly attributed but false announcement about yourself, is not itself "fake" in any way - it's just a lie. There's no question over what it is - an assertion made by you.

But the original example:

"You just need to take a picture of X candidate, stick a made up quote beneath them, and put it on Facebook...

...creeps into "fake" territory because the image is an object with an implicit lie about its origins and motivation. The implicit lie - "I was created in order to raise awareness of a disturbing truth about candidate X" - is distinct from the explicit lie (candidate X said ). Hence the superficially superfluous act of putting the text on a picture, instead of just tweeting "candidate X said ".

The distinction is even more stark with the followup addition: "the fake-ness is even more effective, if the said candidate is the originator and controls the narrative." Here the explicit lie remains identical, but the implicit lie deepens considerably ("look at the fibs the crazy opposition is spreading around").

Re: Twitter to label deepfakes and other deceptive media

#93

Post by celdon25 »

iscrewyou wrote: post #86
celdon25 wrote: post #71

Earlier quoted context omitted.

Will they go as far as to censor statements by politicians such as this? https://dailycaller.com/2016/10/09/media-bias-nbc-news-fact-...

Hopefully. Same as certain individuals yelling innocence while obstructing and tampering jury as well as evidence.

That was meant to be rhetorical, but I forgot that Poe's law was a thing.

Re: Twitter to label deepfakes and other deceptive media

#94

Post by friendlybus »

This problem is in the human mind, there will always be multiple ways to interpret information. It's fundamental. The move to force two different interpretations into one is a cultural and political shift, not a technological move.

The fires of cognitive dissonance and a fixation on hate will be ended. Valid interpretations will be set by a handful of tech companies regardless of the value lost. This is the Thanos snap that has been coming, two halves of the populace into one. Hate is dead, long live the heartless.

Re: Twitter to label deepfakes and other deceptive media

#95

Post by captainredbeard »

kilo_bravo_3 wrote: post #85
kyrieeschaton wrote: post #68

Of course they didn't restrict themselves to "deepfakes"; they made sure to encompass anything "deceptive", "manipulated" media, "tropes", and "unwanted attention" - which, given every video clip has a start and end point, seems tailor made to allow them to flag eg clips of Ralph Northam endorsing infanticide or Hillary Clinton collapsing in public as leading to an incorrect political conclusion rather than being act…

>Ralph Northam endorsing infanticide This is exactly why the policy is needed. If a baby is born with its spinal column outside of its body, a quivering mass of braindead flesh, unsustainable, destined for a slow and lingering death-- it should be terminated. That is infanticide. Humane infanticide. People going around yelling about "Ralph Northam endorsing infanticide" don't care about that. They don't care about th…

Do you have the ability to read minds? Because you're doing some mind reading in that comment.

Many people find the flippant nature of his response disturbing. I'm not "anti-abortion" but I found the Northam situation squicky.

YMMV, and that's the beauty of living in a free and liberal society. Your comment would probably win hearts and minds if it was more about educating folks about the medical reality some parents are faced with instead of castigating and denigrating "the other side" (which doesn't exist, it's a spectrum).

Re: Twitter to label deepfakes and other deceptive media

#96

Post by Bartweiss »

ravenstine wrote: post #73
kyrieeschaton wrote: post #68

Of course they didn't restrict themselves to "deepfakes"; they made sure to encompass anything "deceptive", "manipulated" media, "tropes", and "unwanted attention" - which, given every video clip has a start and end point, seems tailor made to allow them to flag eg clips of Ralph Northam endorsing infanticide or Hillary Clinton collapsing in public as leading to an incorrect political conclusion rather than being act…

They'll label these things "mostly false", a la Snopes.

My first thought was that scope creep from facts to implication is what broke down trust in Snopes.

Per their blog, it appears they have several tiers of action but what standard label. I imagine that the first time people see "false" next to a clip that's real (and flattering to their opinions, perhaps), they're going to discount the warning pretty sharply.

Re: Twitter to label deepfakes and other deceptive media

#97

Post by Bartweiss »

Twitter has an actual post up announcing the policy, with a fair bit more detail on tiers of actions and reasons for actions. (The big one that alarms me is "reduced visibility", which seems far more manipulative than labeling or deletion.)

Link: https://blog.twitter.com/en_us/topics/company/2020/new-appro...

Re: Twitter to label deepfakes and other deceptive media

#98

Post by AndrewUnmuted »

kilo_bravo_3 wrote: post #85
kyrieeschaton wrote: post #68

Of course they didn't restrict themselves to "deepfakes"; they made sure to encompass anything "deceptive", "manipulated" media, "tropes", and "unwanted attention" - which, given every video clip has a start and end point, seems tailor made to allow them to flag eg clips of Ralph Northam endorsing infanticide or Hillary Clinton collapsing in public as leading to an incorrect political conclusion rather than being act…

>Ralph Northam endorsing infanticide This is exactly why the policy is needed. If a baby is born with its spinal column outside of its body, a quivering mass of braindead flesh, unsustainable, destined for a slow and lingering death-- it should be terminated. That is infanticide. Humane infanticide. People going around yelling about "Ralph Northam endorsing infanticide" don't care about that. They don't care about th…

> That is infanticide. Humane infanticide.

> People going around yelling about "Ralph Northam endorsing infanticide" don't care about that. They don't care about the baby, and they CERTAINLY don't care about the mother.

OK, but the problem here is that Ralph Northam, in the video, does endorse infanticide. There is nothing untrue about that analysis. Sure, there may be nothing wrong with that from my or your point of view, but you are tacking on your own narrative to the video all the same. Because you and I don't care about the mother, either, no matter what you may say to the contrary.

That viral video clip is just the VA governor saying some stuff during an interview on a radio program. You and the OP both attached your own critical analysis to that video, and that's great. Platforms which actively discourage this have alterior motives beyond sharing, beyond the truth, and beyond kindness.

You know who cares the least about the mother, out of all of us? Twitter.

Re: Twitter to label deepfakes and other deceptive media

#99

Post by jedieaston »

option wrote: post #65

Yesterday, I searched for a children song from an old soviet movie (my childhood) to play to my kids. Nothing political whatsoever. Youtube promptly showed me the warning for that channel saying that “VGTRK is funded by the Russian government” (which is true). Out of curiosity, I browsed throw tons of (political) clips posted by BBC and did not see a similar warning (regarding UK goverment) or anything close to it.

I’m getting the warning for the BBC on this video: https://youtu.be/dRggIb5Je_w

And I just checked, the CBC gives a similar warning as well.

Re: Twitter to label deepfakes and other deceptive media

#100

Post by iscrewyou »

celdon25 wrote: post #93
iscrewyou wrote: post #86

Earlier quoted context omitted.

Hopefully. Same as certain individuals yelling innocence while obstructing and tampering jury as well as evidence.

That was meant to be rhetorical, but I forgot that Poe's law was a thing.

Looking at the rest of the thread, it was hard to get a rhetorical vibe from your comment. But I get it now.
Post reply on HN