Frances Haugen - Facebook’s Algorithm & Teen Mental Health | The Daily Show — Transcript
Full transcript
- 0:00francis haugen welcome to the daily show
- 0:02thank you for inviting me happy to be
- 0:03here um
- 0:04your name
- 0:06isn't as big as the story that you
- 0:10really i mean leaked to the world
- 0:12you know a lot of people you say francis
- 0:13how can they be like who is that but
- 0:15then if you say to people
- 0:16hey remember how we all learned that
- 0:19facebook and what social media is doing
- 0:21is essentially destroying teenagers
- 0:22brains and it's harming all of us yeah
- 0:24that report happened because of you so
- 0:27let's start with the question of why
- 0:29you've worked for
- 0:31google you've worked for pinterest
- 0:32you've worked for yelp and yet you blew
- 0:34the whistle on facebook why did you feel
- 0:36like oh this this is something i can't
- 0:38sit on
- 0:39so when i joined
- 0:41facebook i thought i was going to work
- 0:42on misinformation in the united states
- 0:45and i was surprised i actually was
- 0:47charged with working on information
- 0:49outside only outside the united states
- 0:51okay and
- 0:52like many people and many technologists
- 0:54i had never really focused on facebook's
- 0:57impact internationally
- 0:58and very rapidly
- 1:02i realized kind of the horrifying
- 1:04magnitude of the danger that we are
- 1:05facing that facebook's algorithms give
- 1:08the most reach to the most extreme and
- 1:10divisive ideas um and that that process
- 1:14is destabilizing some of the most
- 1:16fragile places in the world like
- 1:17ethiopia or what happened in myanmar it
- 1:20you see when when you say that there are
- 1:22some people who will accept it
- 1:23immediately and then there are some
- 1:24people be like that's that's not true at
- 1:26all but i'm sure there are a lot of
- 1:27people who will say well but i mean
- 1:30aren't they always going to be extreme
- 1:31people in the world you know there's
- 1:33always somebody saying something you
- 1:34know facebook themselves say they go
- 1:36like hey we're not we're not part of the
- 1:37problem we're merely a platform that
- 1:40people put their views on we are not
- 1:42part of the problem and you disagree
- 1:43with that so let's imagine you had a
- 1:45relative who had particularly extreme
- 1:48ideas you know
- 1:49we often say like the crazy uncle right
- 1:51right some of us have those uh
- 1:54that person if they're if we're talking
- 1:56to them one-on-one or we're talking to
- 1:57them at a family gathering that scale
- 2:00allows for resolution of ideas okay
- 2:02what's happening on facebook right now
- 2:04is ideas that are the most able to
- 2:06elicit a reaction and the shortest path
- 2:09to a click is anger
- 2:11those get the most reach and ideas that
- 2:13are more moderate or that try to bring
- 2:15us into synthesis or to to help us find
- 2:17a middle path those aren't as likely to
- 2:20elicit a comment from you or elicit a
- 2:22like right and so facebook doesn't get
- 2:24as much distribution so facebook is
- 2:26essentially propelling certain ideas out
- 2:28there so they may say we're the platform
- 2:30but really what they're also doing is
- 2:31they're advocating for certain
- 2:32viewpoints because they push those out
- 2:34because those get people engaged and
- 2:35spinning in the cycle so i think it's
- 2:37the thing where it's not they sit out
- 2:38and go how can we have more extreme
- 2:40ideas i don't think that's what they're
- 2:41trying to do but they do know that there
- 2:43are lots of solutions that are not about
- 2:45censorship it's not about picking good
- 2:47or bad ideas or good or bad people it's
- 2:49about how do we change the dynamics of
- 2:51these systems so that you could have
- 2:54good speech counter counter bad speech
- 2:56how do we give those more of an equal
- 2:58say at the table you know it's
- 2:59interesting that you lay it out like
- 3:00this because i remember talking to my
- 3:02friends about this saying have we become
- 3:05angrier in the world
- 3:07have we become less able to
- 3:09have misunderstandings because i because
- 3:11i sit with my friends and i go like i
- 3:12don't agree on so many things with my
- 3:14friends but i don't remember a time when
- 3:16that meant that i couldn't be friends
- 3:18with them yeah in the world of social
- 3:19media right now it feels like if i
- 3:20disagree with you on this i disagree
- 3:23with you on everything
- 3:24it seems like facebook goes if you
- 3:26disagree on this we're gonna make or
- 3:28we're gonna find more ways to to you
- 3:30know to find where you disagree or how
- 3:32to make you disagree yeah so some people
- 3:33have said this idea of like facebook's
- 3:35always looking for fault lines right so
- 3:37so a high quality piece of content is
- 3:40one that gets lots of engagement and it
- 3:42turns out an angry comment thread
- 3:44where you know it causes us to yell at
- 3:45each other that is viewed as a higher
- 3:47quality piece of content
- 3:49and so i you know during um covet i've
- 3:52had friends who wrote very long
- 3:54thoughtful pieces they went and digested
- 3:56lots of information that's going to
- 3:58cause less knee-jerk reaction than
- 4:00something that really offends you
- 4:03so masks don't work people are clicking
- 4:05on that here's the thing about masks we
- 4:07have to consider in our society boom
- 4:09yeah facebook's not pushing that so
- 4:11what's interesting you know in
- 4:13everything that you've done is even
- 4:14though you're a whistleblower even
- 4:16though you came up with this information
- 4:17you've never said i'm anti-facebook
- 4:19you've never said shut facebook down
- 4:20you've never said this thing needs to go
- 4:21away what you're arguing for is we
- 4:24change our relationship with not just
- 4:26facebook but social media company as a
- 4:27whole
- 4:28explain what that means so often when
- 4:30people talk with this idea if we need to
- 4:31change our relationship with social
- 4:32media it's more about like should we use
- 4:34our phones so much but i'm i'm
- 4:36encouraging us to have a conversation
- 4:37about what's our relationship with
- 4:39companies or what's the company's
- 4:40relationships with us okay facebook
- 4:42knows that they have a level of of um on
- 4:45accountability that's very different
- 4:47than either google or apple or other big
- 4:49platforms because in the case of google
- 4:51we can scrape and download their results
- 4:53and analyze them and see if there are
- 4:55biases in the case of apple we can take
- 4:57apart the phones and put up youtube
- 4:58videos saying apple phones do or don't
- 5:00work the way they claim
- 5:02but in case of facebook
- 5:04researchers and activists have been
- 5:06telling facebook hey we have found
- 5:08all these examples we think there's a
- 5:10pattern here we think there's too much
- 5:11human trafficking we think kids are
- 5:13suffering right and facebook keeps
- 5:14coming back because they know no one can
- 5:16call them on it
- 5:18and saying that's just anecdotal that's
- 5:19not real
- 5:21when you release the information
- 5:22realized it wasn't anecdotal it was real
- 5:23because facebook had done research right
- 5:25you see now this is interesting because
- 5:26this reminds me of a tobacco company yes
- 5:29it reminds us of the you know the fossil
- 5:31fuel companies
- 5:33they do the research they find out
- 5:34something really bad and then i mean
- 5:36obviously they go like we're not going
- 5:37to put this out there but but we know
- 5:39one of the biggest things that's really
- 5:41gotten people worried is how social
- 5:43media affects younger brains i mean it
- 5:46affects my brain i'm i don't consider
- 5:48myself younger anymore but
- 5:50young people are like it seems like the
- 5:52highest at the highest risk
- 5:55yeah facebook's internal research says
- 5:56that the highest rates of problematic
- 5:58use peak at age 14. so you're not
- 6:01allowed to be on the platform until
- 6:02you're 13. so i guess it takes a little
- 6:04bit of time to form that habit but but
- 6:06the largest fraction of any of those age
- 6:08cohorts they sampled and they've done
- 6:09studies that have a hundred thousand
- 6:11people these are not tiny studies
- 6:13they find that the largest fractions say
- 6:16i can't control my usage and i know it's
- 6:18impacting my health my employment or my
- 6:20school
- 6:21is at age 14. and so there's a real
- 6:23thing where young people have both you
- 6:26know they're struggling with issues in
- 6:27their life yes and they don't have the
- 6:29self-control and they say this they say
- 6:31this to researchers they say i know this
- 6:33is making me feel bad and i can't stop i
- 6:36fear i'll be ostracized if i leave
- 6:38um and those factors mean it's not easy
- 6:41for kids just to walk away it's also not
- 6:43easy for people to understand how to um
- 6:45how to fix it you have lawmakers now who
- 6:47are trying to figure this thing out
- 6:49silicon valley is way ahead of them
- 6:51right like way ahead of them lawmakers
- 6:53you see these hearings they'll be like
- 6:54what if what if my facebook is a insta
- 6:56flap they don't understand what's
- 6:58happening so how do we begin
- 7:01creating a world where we're not
- 7:03destroying the companies but we're we're
- 7:05regulating them the way they should be
- 7:07so the main thing i'm advocating for is
- 7:09around transparency because one of the
- 7:11problems today is because facebook is
- 7:12the only one that holds the cards like
- 7:14they can see they can see whether or not
- 7:15they're holding you know a royal flush
- 7:18or not right what they claim um we need
- 7:20to move to a world where we have more
- 7:21access to data and that can be aggregate
- 7:23data it can be privacy sensitive data it
- 7:26doesn't have to you know expose people
- 7:27it's not it's a false choice of privacy
- 7:29versus transparency right and once we
- 7:31have the ability to have conversations
- 7:33we can stop talking about boogeyman
- 7:35social media i see and we can start
- 7:37having conversations on how do we solve
- 7:38these problems i think what we what we
- 7:40need to do and this is what i'm going to
- 7:41spend 2022 doing is we need to start
- 7:43organizing people like i want to plant a
- 7:46youth directed movement where we can
- 7:48begin putting pressure on facebook to
- 7:50release this information because i don't
- 7:52think facebook is just going to do it
- 7:53automatically out of the goodness of the
- 7:55heart we have to force them to
- 7:57and i think there's lots lots of
- 7:58opportunities where we can begin putting
- 8:00pressure on them either socially or
- 8:01financially as a user of social media as
- 8:04a person who speaks to all my friends on
- 8:05social media i have my enemies on social
- 8:07media i appreciate this conversation
- 8:09because i think um if we're not careful
- 8:11we end up in a world where we're all
- 8:13fighting
- 8:14really just because an algorithm is
- 8:16trying to make us fight half of those
- 8:17fights i've noticed even even with the
- 8:19show if i say a thing
- 8:22on the show
- 8:24this has been like the most interesting
- 8:25thing for me is depending on
- 8:27what i say facebook sends that to
- 8:29different people one time we had it on
- 8:31the show
- 8:32one yeah one time we had on the show
- 8:33where um
- 8:34literally i'd laid out both arguments i
- 8:36said here's here's one way to see it
- 8:38there's another way to see it but then
- 8:40on facebook only one way went to each
- 8:42group yeah and i was i was fascinated by
- 8:45that because then people like why didn't
- 8:46you say the other ones like but that's
- 8:47exactly what i said they're like oh well
- 8:48i didn't see that
- 8:50and then you realize you know so one
- 8:52thing that happens is um so let's say
- 8:54you take a video clip so let's say you
- 8:56have out of any given on your shows
- 8:58people go and put on youtube lots of
- 8:59different little chunks
- 9:01people can reseed those onto facebook
- 9:03over and over again they do right and
- 9:05each time that enters into facebook you
- 9:07know someone posted to a group someone
- 9:08doesn't know this it enters into the
- 9:10network in a different way
- 9:12and the algorithm begins picking up data
- 9:14on that and it says oh interesting like
- 9:16these kinds of people engage these kind
- 9:18of stone and so and because it's
- 9:20entering from so many different points
- 9:21it actually gets a chance to learn what
- 9:23kinds of people or communities it
- 9:24engages the most with and so that echo
- 9:27chamber is real like the algorithm
- 9:29pulls us towards homogeneity it's almost
- 9:32like
- 9:33facebook knows that you react most to
- 9:36your neighbor when they play loud music
- 9:39there's no reaction from you when your
- 9:40neighbor's doing other things but when
- 9:42your neighbor plays in doubt music
- 9:43that's when you react actually then
- 9:44facebook goes i'm just gonna it's even
- 9:45worse than that actually so how can it
- 9:47be worse let's imagine there's music
- 9:49that you like and there's music you
- 9:50don't like wow it actually knows you
- 9:53really don't like i don't know reggaeton
- 9:54or something oh wow and one thing is i
- 9:56think it's really sad is um one of the
- 9:57things that was in the documents is that
- 9:59people talk about um especially people
- 10:01from marginalized communities that they
- 10:04will go and and correct people who are
- 10:06spreading um you know racist sexist
- 10:10homophobic comments
- 10:11and guess what now facebook knows you
- 10:13engage with those keywords so they send
- 10:15you so they send one
- 10:16yeah i i i didn't know they were doing
- 10:18that they don't do it on purpose it's
- 10:20it's a it's a side effect it's a side
- 10:22effect of the way the algorithm works
- 10:24so the algorithm is designed to get as
- 10:25much engagement as possible and the
- 10:27engagement unfortunately you're more
- 10:29likely there's a there's a car accident
- 10:31i'm going to engage yeah yeah or or more
- 10:33of um if they had an option of showing
- 10:35you a stream of different things they
- 10:37know you gawk a car accident so they
- 10:38show you more
- 10:40man yeah this is hard because on the one
- 10:42hand i go like well i need to stop
- 10:43looking at car accidents but the brain
- 10:44is the brain it's the brain you know if
- 10:47if all the um ai is not intelligent
- 10:50right we like to say artificial
- 10:51intelligence yes but but people who
- 10:53actually study it call it machine
- 10:54learning because it's not intelligent
- 10:56it's just a hill climber it's optimizing
- 10:59and uh you know think of all the
- 11:00different pieces of content you could
- 11:02expose to
- 11:04it it's not trying to show you most
- 11:06extreme content it happens to be to
- 11:08fulfill its goal function
- 11:09it mindlessly pushes you towards it and
- 11:12so this is about facebook making choices
- 11:14to not fix it that facebook has lots and
- 11:17lots of options that don't deal with
- 11:18letting the machine run wild they're
- 11:19allowing the machine real well
- 11:22money money that's what it is
- 11:25thank you so much for being here this
- 11:26has really been enlightening good luck
- 11:27on the rest of your journey thank you so
- 11:29much
- 11:30if you want to learn more about
- 11:31francis's advocacy work then you can go
- 11:33to francis haugen.com
- 11:38[Music]
- 11:47you
About this transcript
This page contains the full transcript of Frances Haugen - Facebook’s Algorithm & Teen Mental Health | The Daily Show by The Daily Show, generated from the public captions YouTube serves with the video. The transcript has 2,502 words across 374 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.
What you can do with it
Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.
Free YouTube transcript tool
YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.