0:15

Pushkin, I'm

0:19

may have Higgins, and this is Solvable

0:22

Interviews with the world's most innovative

0:24

thinkers who are working to solve the

0:26

world's biggest problems. In

0:29

this episode, Anne Applebaum is in conversation

0:32

with researcher and data analyst Renee

0:34

Deresta about her solvable, which

0:37

is the growing spread of dangerous

0:39

misinformation online, especially

0:42

on social media. For the

0:44

most solvable, I think we need increasing

0:46

awareness, increasing cooperations,

0:49

helping algorithms make better decisions, recognizing

0:52

that recommendation engines

0:55

are not functioning as they should, and that we should

0:57

be taking tangible steps to think about

0:59

ways in which algorithm curation serves

1:02

information to people. In

1:05

late twenty sixteen, Oxford Dictionaries

1:07

selected post truth as their

1:09

word of the year, defining it as

1:11

relating to or denoting circumstances

1:14

in which objective facts are less

1:16

influential in shaping public

1:19

opinion than appeals to emotion

1:21

and personal belief. It's

1:23

like, I want to believe that nachos

1:25

are the ideal balanced nutritional

1:28

snack that appeals to my emotional

1:30

and personal belief system, because God,

1:32

I love nattos. So

1:34

I'll go and I'll find some vague chitchat

1:36

online that tells me, you know, something

1:39

melted cheese is totally full of calcium

1:41

that is good for your bones, and it's important

1:44

for you, as an immigrant to the US, to assimilate

1:46

by eating their national dish of natos.

1:50

So I'll convince myself of

1:52

that, and I'll maybe even eat myself

1:54

into a delicious early grave. The

1:58

age we live in, the digital age,

2:00

affects every narrative we see and absorb,

2:03

and that can be news based, or cultural

2:05

or artistic. We have always

2:07

had an instinct to find information that

2:09

sinks with our perspective, and

2:12

now a host of new platforms

2:14

are only too happy to oblige

2:16

that part of us. Pew reports

2:18

that an analysis of almost

2:21

four hundred million Facebook users interactions

2:23

with over nine hundred news outlets

2:26

found that people tend to seek information

2:29

that aligns with their views.

2:31

That makes many of us vulnerable to accepting

2:33

and acting on misinformation. Social

2:37

media firms are under pressure to halt the

2:39

spread of fake contents on their platforms,

2:42

and we know that the problem has both

2:44

human and technical side, and

2:47

so too does any potential solution.

2:50

Reneed Arresta is the director of research

2:53

at New Knowledge and a Mozilla

2:55

Fellow in Media Misinformation and Trust.

2:57

She investigates the spread of malign narratives

3:00

across social networks and helps policymakers

3:03

to understand and respond to the problem.

3:06

Renee has advised Congress and the State Department,

3:09

and she studies some fascinating areas

3:11

of disinformation in contexts

3:14

like pseudoscience, conspiracies,

3:17

terrorism, and state sponsored

3:19

information warfare, all

3:21

that spooky stuff. I'm so glad she is

3:23

scouting ahead and sending us back

3:26

the best ways to deal with this. Let's

3:28

take a listen and I'll speak to you after.

3:31

So, Renee, you're one of the few people

3:34

who identified the problem

3:36

of online anti

3:38

vax disinformation very

3:41

early on. How did you first come into contact with

3:43

the problem? How did you know it was a problem at all. I

3:45

started working on a law in California

3:47

called SP two seventy seven, and it was a law to

3:49

eliminate vaccine opt outs. And I was

3:51

a parent at a new baby, and I

3:53

wanted, as a mom, just to volunteer

3:55

to help get this law passed. So I am

3:58

a data analyst, and I offered to do

4:00

some analysis into things like the

4:03

social media conversation around the law. And

4:05

I was really surprised because the legislators,

4:07

there are a number of legislators on both

4:09

parties who were supporting the law. They were saying that

4:11

their constituents were polling at around eighty five

4:14

percent in favor, but the social media

4:16

conversation was almost one hundred percent negative,

4:18

and that was on Facebook and Twitter. So

4:20

I started working with another data scientist named

4:22

Glad Latan to look at

4:25

the conversation on Twitter, to look at the different

4:27

distinct groups, how they were evolving their messages,

4:30

how they were connecting with other activists

4:32

outside of California, how sometimes activists

4:34

outside of California. It turned out we're pretending

4:37

to be Californians, bunches

4:39

of new accounts that had been created, and we

4:41

were really looking at the idea of what had no

4:43

name then but kind of came to be called manufactured

4:45

consensus, the idea that the conversation

4:48

online was really being driven by a

4:50

relatively small number of people who

4:52

were using things like tools to be always

4:54

on, constantly being in the hashtag

4:57

Facebook groups and ads to amplify their

4:59

message, and then the way that the algorithm

5:02

was amplifying the message. In addition

5:04

to that, so ways in which I, as

5:06

a parent who had just gotten involved in the conversation,

5:08

had just demonstrated an interest in vaccine

5:10

policy, was all of a sudden getting pushed tons

5:13

of anti vaccine content on Facebook.

5:15

It was recommending groups to me, it was recommending

5:17

pages to me. And the realization

5:20

that what was really not

5:22

a very large number of people was actually

5:25

having an extreme disproportionate

5:27

amount of a share

5:29

of voice in the conversation, and

5:32

did you have to create tools in order

5:34

to begin identifying who the people were

5:36

who was being pushed. They were actually not very

5:38

quiet about that. There was a page called tweet for

5:41

Vaccine Freedom, and it was actually you know,

5:43

when out of state activists were asking how can

5:45

we help because the entire anti vaccine movement across

5:47

the entire United States decided to fight

5:50

this battle. They would say like, oh, you should

5:52

just create an account and say or from California.

5:54

So it was actually really transparent. It wasn't that hard

5:56

to figure out that there were people pretending

5:58

to be from California. There were also Twitter

6:00

accounts that all of a sudden had a vested interest

6:03

in California politics. But if you read their past

6:05

material, which again is also public it was really

6:07

right out there that that's not where they were actually from. Kind

6:09

of a very interesting because it was extremely

6:12

small, local and niche we you know,

6:14

we thought in California. But as the law

6:16

began to get more press coverage and things

6:18

or would actually be like comments section

6:20

battles, you know, the same kinds of stuff that we saw

6:22

later with you know, entities that go and are

6:25

like almost incentivized to leave

6:27

comments on news articles to shape a perception

6:30

about the topic. And actually we on the

6:32

provac side thought, oh boy, I guess we're gonna

6:34

you know, we need to do this too. Are we

6:36

really engaging in Okay? They commented

6:38

over here, so you know, we have to go comment over here. They

6:40

have bots that are on twenty four to

6:43

seven? Do we need bots that are on twenty four seven?

6:45

Just became this this interesting firsthand experience

6:47

of what it was going to be like to try to run

6:49

any kind of influence or policy campaign in the

6:51

future. I found it really troubling, especially when

6:54

the algorithms just began recommending anti

6:56

vaccine content to me constantly. And

6:58

how did the Facebook and Twitter and other algorithms

7:01

work. Were they affected by this campaign? The

7:03

where the search engines affected by it. I don't think

7:06

the search engines as much because it was

7:08

you know, the Google is a little bit more sophisticated

7:10

about this stuff than the social platforms. Social

7:13

platforms the number one signal that they're using as

7:15

popularity, and so you either

7:17

if you have real popularity or if you can feign

7:19

popularity. The number of likes and

7:21

engagements and comments and things is what decides,

7:24

you know, whether this is how Facebook was deciding

7:26

what gets pushed into your feed. Instagram

7:28

is like that too. Google has

7:31

a framework now it has a proper

7:33

name. It's called Your Money or Your Life, and it

7:35

says that on topics related to

7:38

health issues and financial issues they have to have

7:40

a higher standard of care to make sure that it

7:42

isn't just what's popular that's rising to the top.

7:45

But even with that policy, one of

7:47

the things that we consistently see is anti

7:49

vaccine activists producing content

7:52

at a higher rate and also candidly

7:54

more engaging content, you know, a much more emotionally

7:56

resonant versus more authoritative

7:59

medical quote unquote establishment doctors,

8:01

the CDC, the National Instituites for Health,

8:03

their contents not as emotionally resonant. It doesn't

8:05

get as much engagement, and so the search

8:08

engines and the algorithm aren't amplifying

8:10

the more factual, reality

8:13

based content, and instead what we're getting is this conspiratorial

8:15

stuff. Walk me through what it means

8:17

to be emotionally resonant online. Is

8:19

this something that's being done deliberately to the people who

8:21

are creating it understand that that's what it is?

8:24

Or is it that the human brain is just tuned

8:26

to conspiracies and prefers them. Some

8:28

of it is platform culturism, of it

8:31

is the way that the algorithm understands

8:33

engagement. So there's the human element which gets

8:36

kind of the initial signal shows that

8:38

there's a lot of people who are watching this, and the algorithm

8:40

recognizes that a lot of people are watching it and

8:42

then begins the amplification process. But

8:44

the first step is actually the content, of course, and

8:47

in that particular area, it's usually

8:49

a first person, you know, looking directly

8:51

at a camera, speaking about a personal

8:54

experience they've had, recounting a

8:56

narrative or an interesting story. So

8:58

a lot of times with the anti vaccine movement, that's a

9:00

person claiming that their child has autism

9:02

and telling a story, you know, usually

9:04

very sad story about their child's

9:06

health, and so it is engaging.

9:09

It is much more resonant versus

9:11

seeing kind of infomercial about how

9:13

vaccines don't cause autism because thousands

9:16

and thousands and thousands of studies have said that they do

9:18

not. I know that you were part of the Senate

9:20

commission that looked through material

9:22

that we knew that Facebook handed over to Congress

9:25

which was originally created by the IRA, the Russian

9:28

Internet Agency, in order to influence the

9:30

US elections. When you looked over that material,

9:33

did it seem to use those same tactics?

9:35

Can you see a relationship between the way the

9:37

Russian influence campaign worked and the anti

9:39

vax campaigns? The Russian content was distinct,

9:41

and that this was a foreign intelligent

9:44

service of a foreign entity that was trying

9:46

to pretend to be American. So it

9:48

was far more duplicitous than anything that we've

9:50

seen related to domestic

9:53

activists pushing for a cause. Really, but

9:55

what was happening there was again they were taking these

9:57

extremely big topics

9:59

things like who is America for? What

10:01

does it mean to be an American? How do we feel about immigration?

10:04

How do we feel about gay rights? How do we feel about

10:06

police brutality? They

10:09

were creating these pages, and each page was

10:11

designed for a very particular type of

10:14

person, So they were really creating these tribes,

10:16

again relying on the sort of first person experience,

10:18

first person concerns and fears,

10:21

and putting out content that was again very

10:23

much focused on achieving an emotional response.

10:26

So for the black community, the content took

10:28

the form of constant references

10:30

to police violence mixed in

10:32

with narratives of pride, and so it was

10:34

really very much designed to evoke

10:37

cultural pride and then also a

10:39

sense of deep harm. And on the right

10:42

leaning pages, it was really concerned about

10:44

what America is and who it's for, and so a

10:46

lot of photos of things like homeless veterans.

10:49

This is a very real problem

10:51

that we have in this country, and they were using the images

10:53

of homeless veterans to say, why are

10:55

we allowing in all of these outsiders when we can't take

10:58

care of our own. This is how propaganda is most

11:00

effective. It's when it has some degree of truth to

11:02

it, and it spins it just enough

11:04

that it doesn't necessarily trigger the

11:07

part of the brain that says, hey, this is false. Instead

11:09

it the person relies on

11:11

the emotional reaction to it, and that's how

11:13

they begin to develop

11:15

a sustained relationship with the page and sustained

11:17

engagement with that type of content. You know, as

11:19

I'm listening to you, I'm wondering whether different

11:22

kinds of propagandists they understand now that

11:24

they need to tailor messages to particular audiences.

11:26

Is it the case that some of the solutions to

11:29

this they're also going to involve thinking

11:31

differently about different audiences or offering

11:33

different kinds of counter messaging or counter

11:36

strategies to different audiences. Yeah. Absolutely,

11:38

And this is something that you know. A third area

11:40

I worked in was countering violent extremism. Briefly

11:43

was Isis. The idea that we would kick Isis off

11:45

the platforms was sort of a stretch at the time.

11:47

There were a lot of people who were very concerned

11:50

at the idea that we would delete terrorists accounts,

11:52

and so a lot of the focus instead was on counter

11:54

messaging. How do we reach these audiences that

11:56

are receptive to ISIS propaganda and present

11:59

counter narratives to them. Who is the authentic

12:01

voice for the counter narrative. It's definitely not

12:03

the United States State department, So who

12:06

is it and what are the ways in which we can

12:08

come together to think about ways to counter

12:10

message to try to present people

12:13

with it an alternate, also emotionally

12:15

resonant narrative instead of just saying it's

12:17

a bad idea to be a terrorist because you're going

12:19

to go to jail or you're going to die. A lot of the tribal

12:22

deep affinity ties is what is my place

12:25

in society? This is something that comes up with conspiracy

12:27

theorists also, they're looking for

12:29

answers, they're looking for an explanation. What

12:32

you get hooked into oftentimes is what

12:34

is most visible to you, what's most prevalent

12:37

in your space at that moment.

12:39

Now that we're spending so much more of our time online,

12:42

things like ad targeting and

12:45

participation in Facebook groups where

12:47

you're kind of declaring a particular alignment

12:50

mean that bad actors who want

12:52

to target you with certain types of propaganda can find

12:55

you very easily. And can we reapply

12:57

some of that thinking back, for example, to

13:00

the anti vax problem. Can we think

13:02

about counter messaging there? Can we think about

13:05

how to reach people using counter

13:07

emotional stories? Yes, absolutely, that

13:09

is That's something that our groups

13:11

like Voices for Vaccines are trying to work

13:13

on that. The group that was formerly called every child

13:16

by two, it now goes by vaccinate your family is trying

13:18

to do that. We have to get out of statistics

13:20

and get into storytelling. That's

13:23

the one of the key takeaways of how the

13:25

information ecosystem has evolved. If

13:27

you look at even just from a design perspective, one

13:29

of the things I always get at is the the

13:32

subject of the narrative is

13:34

interesting when you're thinking about how to counter message

13:36

to a particular group of people. But when you think

13:38

about this as a problem written large, a

13:40

lot of it comes down to the algorithms and

13:42

the design. And so memes

13:45

in particular are getting more and more important

13:47

in our lives. And that's because the design

13:49

of the platform itself is privileging

13:51

this large you know, this large square image

13:54

or this piece of video, this short video

13:56

clip. So what can you convey in

13:59

the construct of that design.

14:01

As people are scrolling by, they see your

14:03

message in it immediately sticks. The fact

14:06

that the algorithm will continue to serve up

14:08

types of content that you've engaged within the past

14:10

means that if you do engage with anti vaccine

14:13

content, you're likely to see more of it. The

14:15

challenge of algorithms

14:18

that don't know what they're pushing because they

14:20

have no actual awareness of what the underlying content

14:22

is. So they treat something that's potentially

14:24

radical, they treat something that's potentially blatantly

14:28

false the exact same way that they would treat something

14:30

that's accurate or uplifting. They don't

14:32

actually know. They just know that this content

14:35

drives engagement, and so they continue to show

14:37

it to people. We see all this disinformation

14:40

online. We you know, we hear about it.

14:42

You know, we can sometimes see it in our Google searches.

14:44

But doesn't really matter. I mean, for example, in

14:47

the anti vax campaign, has this really

14:49

affected anything, Does it make any difference? Or is

14:51

this just stuff that exists somewhere in the ether and

14:53

if we ignore it will go away. Let me give

14:56

you two quick examples on that. First of all, with the anti

14:58

vaccine movement, Yes, it absolutely has an impact.

15:00

It really creates a lot of fear and hesitancy,

15:03

and that translates very directly into

15:05

vaccination rates declining in the communities

15:08

that are that are seeing it. And so this is

15:10

something that in California. The reason I

15:12

started looking at it was because immunization

15:15

rates in California communities had declined, and when

15:17

I was trying to find a preschool for my son,

15:20

I was actually looking at these rates, and there

15:22

are certain schools in California with thirty percent

15:24

immunization rates, which is terrifying.

15:27

That's like South Sudan. The

15:29

reason that we passed the lawn California was because we

15:31

wound up with the Disneyland measles outbreak, where

15:33

two hundred and something people got sick and I believe

15:36

a quarter had to be hospitalized. So

15:38

this was a very real outcome of

15:41

that kind of misinformation becoming so

15:43

pervasive to people, creating

15:45

that very real fear and then leading to an

15:48

outbreak in the case of Russia.

15:50

Just because a lot of people think about this is just related

15:52

to the election. No, what they were doing was they were also

15:54

creating real world events. So they were

15:57

sponsoring protests, and one

15:59

of the things that they sponsored was an incident in

16:01

Texas where they had two competing

16:03

protests on the same day at the same time. So

16:06

from Saint Petersburg, Troll created

16:08

a Facebook event saying that people

16:10

with Texas Pride had to come and protest outside

16:13

of an Islamic center to defend their way of life.

16:16

They also posted an event calling

16:18

on members of the Islamic Center to come out and

16:20

defend the Islamic faith. So they

16:22

sponsored two protests on the same day

16:24

at the same time, across the street from each other.

16:27

And you can go on YouTube and you can see the video

16:29

footage from that day of people showing

16:31

up with kind of anti Islamic

16:33

material on one side of the street and then people on the

16:35

other side of the streets screaming back at them, and police

16:38

getting involved in breaking up altercations.

16:41

So this is an example of very

16:43

real world tension erupting

16:45

as a result of online disinformation.

16:48

When you first started looking at this problem, did

16:51

people believe it was a problem. Opinion

16:53

polls all showed people were in favor of vaccinations.

16:56

You saw something quite different online. How

16:58

did you convince people that this was something they need to take

17:00

seriously. In the California case in

17:02

particular, I sent what

17:04

I was seeing, you know, kind of quantifiable

17:07

evidence to the legislators

17:09

and said, I don't think that people

17:11

are screaming at you online, they're threatening you online,

17:14

You're seeing all of this anger and rage

17:16

in the hashtags, I don't think that

17:18

these are not your constituents, where it's pretty

17:21

pretty clear that these are not all even

17:24

Californians. So when you make your decision,

17:27

I would lean into the polling

17:29

numbers and the communications with your actual constituents.

17:31

I don't think that we can treat the online conversation

17:34

as representative of the reality of

17:36

the population of California. So

17:38

in that particular case, it

17:40

was just really kind of appealing directly to

17:42

the legislators with the evidence the challenges

17:45

it really does bump up against things like freedom

17:47

of expression right. So you have a right to

17:49

have an anti vaccine opinion. Of course you have

17:51

a right to put the content online. The

17:54

challenge was at the time, the

17:56

recommendation engine, the trending algorithm,

17:59

the ways in which Twitter

18:01

and Facebook were amplifying information

18:04

was very different, far

18:07

more primitive then than it is even now two

18:09

and a half years later. After those of us who

18:11

work on this challenge have kind of been constantly beating

18:13

the drum with example after example

18:15

of example of how this is manifesting

18:18

in the real world. How do we preserve

18:20

freedom of expression while at the same time recognizing

18:23

that the platform is

18:25

pushing this point of view at people people

18:27

aren't even know me. In particular, I'm not going

18:30

out there typing in anti vaccine search

18:32

terms. The recommendation engine is just

18:34

pushing it to me because it's seeing that I've expressed

18:36

an interest in vaccines in general.

18:39

As part of working on this law, I suppose

18:41

there's also a question of Okay, you have a right

18:43

to write something, but then do you have a right

18:46

to artificially amplify it using

18:48

bots and search engine optimization?

18:51

So everyone has the right to freedom

18:53

of expression online. The secondary

18:56

piece of that, though, is do you have a right to free

18:58

reach your right to algorithmic

19:00

amplification. Nobody has that right.

19:02

That is not part of the First Amendment, That is not

19:05

part of our cultural experience

19:07

of what it means to have a right to express, or you

19:09

have never had the right to free

19:12

mass dissemination as well. That's

19:14

the piece where as people begin

19:16

to talk about how the platform should think about

19:18

these things. One of the ways that we can continue to

19:20

maximize freedom of expression is

19:23

to allow people to speak, but also for

19:25

the algorithm to perhaps not

19:27

begin to take that kind of

19:30

sensationalist content and

19:32

proactively broadcast

19:35

it out to massive quantities

19:37

of people because it checks

19:39

the boxes of being sensational and emotional.

19:42

And do you think that it's going to be enough

19:45

to discuss this with the platforms, for people

19:47

like you who have you, who are respected

19:49

on these issues, to talk about it with people at Facebook

19:52

and Google, or is this something that we're

19:54

going to need to regulate or have Congress

19:56

step in on. I don't think you can

19:58

have Congress regulate what

20:00

algorithms amplify. I think that that would probably

20:02

be a little bit too close to Congress

20:05

making decisions on speech. A

20:07

lot of the dissemination that come about

20:09

through inauthentic amplification

20:12

through things like bots and stuff, can be addressed

20:14

without even knowing what the narrative is actually about.

20:16

So you're not looking for content related

20:18

to a particular topic. You're looking for particular

20:21

dissemination patterns. So you're looking

20:23

at the authenticity of the accounts. Are these

20:25

real accounts where they all created yesterday?

20:28

Are they bots? Are they majority automated?

20:30

Are they Twitter does have now a designation

20:32

of something that considers a low quality account

20:35

ways in which it surfaces

20:37

top tweets, as opposed to just

20:40

the straight up reverse chronological order where

20:42

you see every single tweet about a particular

20:44

hashtag, giving the user some control.

20:46

So people who do want to go see that kind

20:48

of fire hose of every single tweet

20:51

coming through about a topic can go and do that. But

20:54

the majority of people who just want to get

20:56

the kind of quick takeaways are seeing more

20:59

kind of higher caliber content. And

21:01

that sounds like you do believe algorithms

21:03

could eventually identify

21:06

quality content that they could encompass

21:08

a notion of better or more comprehensive

21:12

or more fact based. Remember

21:14

the olden days of the Internet

21:16

where you had email spam, right, we did build

21:19

classifiers, We did build tools

21:21

to think about how to ensure that crap

21:24

wasn't flooding people's inboxes, that there

21:27

wasn't this mass cognitive load

21:29

every time you opened your inbox of having to sift

21:31

through all of the garbage to find the communications

21:34

from people that you actually wanted or or find

21:36

the things that were really intended for you. We

21:38

need to put some things in place here to

21:41

improve the system, to improve the user

21:43

experience, to improve the outcomes. There

21:45

were things like recognizing that certain

21:47

domains were just not reputable

21:50

domains that most people wanted in their inbox, and

21:52

so some of this was user filtering, you know, feedback.

21:55

You remember you used to kind of mark things as spam much more

21:57

regularly then. It didn't mean that there were never

21:59

false positives. There are still false positives today,

22:01

But it was how can we create greatest

22:05

value while at the same time recognizing

22:07

that there are extremely

22:10

coordinated, deliberate groups of people

22:12

working to manipulate and evade that detection,

22:14

in the case of spam, to wind up in your inbox

22:17

and in the case of social algorithmic manipulation

22:19

to wind up in your feed People who are concerned

22:21

about this problem, people who worry about online

22:23

disinformation, people who worry they're getting bad

22:25

information. Is there anything they

22:28

can do about it? Is there something that ordinary

22:30

people can do to fight back? Stopping

22:32

this spread a lot of the time is something where

22:35

individuals really have a lot of power. It's

22:38

been for a long time, you know, kind of a cultural

22:40

norm where if you see someone sharing something

22:42

a little bit nutty to just kind of ignore it, just

22:44

let it go by. I don't think that

22:46

that's necessarily really helped us. I've tried

22:48

lately to try, like commenting gently

22:51

or sending a private message saying hey, I don't

22:53

think this is necessarily the most reputable source.

22:55

Maybe you know, here's a fact check

22:57

on that. There's a lot of evidence that says that interventions

23:00

from people, you know, even

23:03

in the kind of counter radicalization space,

23:05

that really engagement with friends

23:07

and family and people were there's a base of

23:09

trust and an assumption of goodwill. People

23:12

are receptive to rethinking maybe

23:15

why they chose to share something. And then when

23:17

you see something that makes you feel highly emotional

23:19

and you go to click the share button or the retweet button

23:21

just because you know, you feel outraged

23:24

and you need to tell the world, that's where I think

23:26

taking the extra second to stop and do the fact

23:28

check, to stop and see is this a reputable

23:31

domain or a reputable account, it really

23:33

makes a difference. So friends, don't let

23:35

friends share disinformation, and

23:38

always check whose account you're

23:40

retweeting or reposting before you do it.

23:42

Yeah, I mean, I've made this mistake a couple of times.

23:44

I remember I once retweeted something

23:46

and a friend of mine ping me and said, hey,

23:48

I think you should go read the rest of that accounts

23:51

tweets, And I went and looked,

23:53

and I ninety nine percent sure it was a bot,

23:55

and I was like, oh, I fell

23:57

for it, you know so.

24:00

But that's the kind of thing where it's far

24:03

better to tell somebody. I mean, you can just unretweet,

24:05

you just click the button again. And it's more challenging

24:07

if you are a person with a very, very large following,

24:09

and it usually helps to send a follow up or something

24:11

and say, hey, I inadvertently spread some misinformation.

24:14

It's come to my attention that this is not real, or

24:17

here's the actual story.

24:19

It's so wild to hear about these

24:21

disinformation campaigns online right

24:24

now because here in the US there have been

24:26

eight hundred and eighteen measles cases reported

24:29

in this year's outbreak. It's already

24:31

the largest since nineteen ninety four.

24:33

People are in hospital here because

24:36

of misinformation, and New

24:38

York is seeing the fastest spread, particularly

24:41

in Orthodox Jewish communities. The

24:43

thing is that in that specific case,

24:45

the misinformation about vaccines was

24:47

not spread online, but through physical

24:50

handbooks and phone conferences. The

24:52

internet amplifies what we already

24:55

do, so changing algorithms

24:57

and policies and our own behavior online.

25:00

It's all going to take a lot of changing,

25:02

and I'm really grateful to people like

25:04

Renee who work towards that

25:06

every day.

25:11

Solvable is a collaboration between Pushkin

25:14

Industries and the Rockefella Foundation,

25:16

with production by Chalk and Blade. Pushkin's

25:19

executive producer is Mia LaBelle. Engineering

25:22

by Jason Gambrell and the fine folks

25:24

at GSI Studios. Original

25:27

music composed by Pascal Wise. Special

25:30

thanks to Maggie Taylor, Heather Fame,

25:32

Julia Barton, Carlie Migliori, Sherif

25:35

Vincent, Jacob Weisberg, and Malcolm

25:37

Gladwell. You can learn more about solving

25:40

today's biggest problems at Rockefella

25:42

Foundation dot org, slash solvable.

25:45

I'm Mave Higgins. Now go solve

25:48

it.