[00:03.150 --> 00:08.290]  Hi, I'm Leo, the founder and CEO over at Graphistry, and I'm going to be talking today about Project
[00:08.290 --> 00:13.630]  Domino, which is not even at Graphistry, but a community open citizen data science effort
[00:13.630 --> 00:18.610]  against COVID misinformation. And before we go into it, I do want to say thank you to
[00:18.610 --> 00:24.890]  the DEF CON staff, AI Village staff, and the Project Domino volunteers, and also AI Village
[00:24.890 --> 00:30.630]  in particular, Sven. This is a crazy time for all of us, and so to be able to put this together,
[00:31.570 --> 00:36.510]  a big thank you for that. I'm honored to be giving this as a keynote. I'm hoping folks
[00:36.510 --> 00:41.530]  here will benefit from hearing what we've been learning, and also if you are impacted somehow
[00:41.530 --> 00:49.070]  by COVID and want a way to give back, this might actually be an opportunity for you as well.
[00:49.090 --> 00:55.790]  With that, I want to start with an example. This is about three months ago. It might feel like
[00:55.790 --> 01:01.530]  three years for some of us. And around the time when the Seattle breakout was happening,
[01:01.530 --> 01:05.130]  our friends were kind of dealing with the hospital situations over there.
[01:05.570 --> 01:12.390]  I think it was already like people were actually dying already in Seattle. Governor Kevin Stitt
[01:12.390 --> 01:17.850]  from Oklahoma posted this strange tweet, eating with my kids and all my fellow Oklahomans at the
[01:17.850 --> 01:23.710]  collective OKC. It's packed tonight, support local, Oklahoma proud, basically. And then he clarified
[01:23.710 --> 01:29.870]  next day in a press release, we encourage Oklahomans to go out to restaurants. So for us,
[01:29.870 --> 01:37.910]  with ties to the medical system, our head kind of was exploding. Like, what's up? Fast forward to a
[01:37.910 --> 01:46.290]  couple weeks ago, and unfortunately for Governor Kevin Stitt and his family, he tested positive
[01:46.290 --> 01:51.610]  for COVID. And unfortunately for his constituents who are relying on him, they have about a thousand
[01:51.610 --> 01:56.370]  per day infections at this point, because kind of what we feared would happen is exactly what
[01:56.370 --> 02:05.710]  happens. And now Oklahomans are dying. So it's... I do want to do kind of some disclosures and stuff
[02:05.710 --> 02:10.090]  in a bit. But to us, it's not about a left or right thing. It's like we're, you know,
[02:10.090 --> 02:16.790]  this is a medical health emergency. We need to do stuff. And that's kind of a good note,
[02:16.790 --> 02:24.750]  kind of where this project came from. We're kind of looking at kind of what can we do.
[02:24.930 --> 02:30.510]  And there's this interesting statement by the... from the World Health Organization saying,
[02:30.510 --> 02:35.910]  we're not just fighting an epidemic, we're fighting an infodemic. And kind of the reasoning
[02:35.910 --> 02:43.510]  that we kind of believe in here is until there's a medical cure, a pill, a vaccine,
[02:43.510 --> 02:52.470]  been tested, it's safe, kind of the only thing we really have is behavior change. So for that,
[02:52.470 --> 02:58.810]  unfortunately, you either need to get top-down policy enforcement or bottom-up
[02:59.510 --> 03:04.450]  kind of citizen voluntary compliance. And unfortunately, we have folks standing in the
[03:04.450 --> 03:09.450]  way of that. And, you know, for Seattle, for Oklahoma, for all of us at home,
[03:09.450 --> 03:14.450]  we're seeing kind of the effects of failures to win that fight.
[03:15.390 --> 03:21.270]  Some quick disclosures here. We are going to talk about real people. We've published... actually,
[03:21.270 --> 03:26.110]  we haven't directly published. We publish through partners. So we empower researchers to do through
[03:26.110 --> 03:30.250]  our tech. So they've published on some of this stuff before. And based on some of the results
[03:30.250 --> 03:38.070]  today, I expect to see some accounts going out and hopefully some takedowns. Likewise,
[03:38.070 --> 03:44.650]  this is a volunteer organization. So the views expressed by us, me, this talk, they're not
[03:44.650 --> 03:51.890]  necessarily representative of the employers of the volunteers. Likewise, you should expect some
[03:51.890 --> 03:57.230]  bias here. So a lot of us, actually, some people on the team were even affected by COVID and didn't
[03:57.230 --> 04:02.290]  have their lives impacted by that. Family members do. Their work and work life has been impacted.
[04:02.410 --> 04:07.490]  Misinformation is a broader topic. I'm personally in a place some folks may remember called the
[04:07.490 --> 04:12.790]  Soviet Union, where propaganda was a thing and surveillance was a thing. So conflicting thoughts
[04:12.790 --> 04:18.390]  on that. But I do want to point out from the team, I do especially want to call out some of the kind
[04:18.390 --> 04:24.370]  of folks on the leadership team. Sean Griffin's been involved in pandemic response. Actually,
[04:24.370 --> 04:31.010]  he helped both Obama and the Trump administration with pandemic planning. Anita was over at the NSF
[04:31.010 --> 04:36.870]  and looking at a lot of digital crime, blockchain, stuff like that. She's at UIUC now.
[04:36.870 --> 04:42.850]  Cody Webb has been this research powerhouse for our team and researching misinformation
[04:43.530 --> 04:49.410]  forever. I shared the names of some of the engineers helping on our backbone.
[04:50.650 --> 04:55.850]  The first example today will be actually Jeff Goldberg over at Social Forensics and David
[04:55.850 --> 05:01.470]  Knickerbocker, an independent researcher, has been helping us a bunch on. And I think, oh,
[05:01.470 --> 05:06.650]  yeah, and then tomorrow, if this is interesting to you, kind of the same type of stuff we're doing
[05:06.650 --> 05:12.170]  for COVID. Andrea's going to be talking over at the Biohacking Village about imagine you're a
[05:12.170 --> 05:17.590]  cancer patient and you go to Facebook and then you see all the same misinformation giving you
[05:17.590 --> 05:25.510]  false hope or misguiding you. So it's important stuff. Okay. So today, I'm going to talk about
[05:25.510 --> 05:29.950]  basically three things. First, I just want to take you down the rabbit hole through an example
[05:29.950 --> 05:36.850]  from anti-vaccination misinformation. Second, I want to then give you more of a security data
[05:36.850 --> 05:40.170]  science perspective and how we're actually going about it, trying to figure out scale and
[05:40.170 --> 05:45.450]  autoresponse and all that stuff. And then briefly, but incredibly important,
[05:45.450 --> 05:51.790]  why we're doing this as Open Citizen Data Science and why I'm actually pretty hopeful about where
[05:51.790 --> 05:56.470]  this is going long term. And maybe this could kind of give you a chance or give you ways to
[05:56.470 --> 06:07.810]  help out the community. Okay. So let's dive in. So we were doing an investigation and this name
[06:07.810 --> 06:13.810]  popped up, Sherry Tenpenny, and we decided to see where we go for the purposes of this talk.
[06:14.830 --> 06:21.450]  And for some folks, if you're from Australia, this may be a familiar name. She's Wikipedia
[06:21.450 --> 06:28.890]  famous or infamous at this point. So her 2015 lecture was canceled in Australia because
[06:28.890 --> 06:36.510]  apparently she is involved in a lot of vaccine misinformation and private organizations in
[06:36.510 --> 06:46.000]  Australia wanted no part of that. And so they didn't support her. For our purpose and purposes,
[06:47.840 --> 06:52.680]  I do want to point out there's kind of an ecosystem with different incentives here.
[06:52.680 --> 06:59.420]  And so if we look at somebody like Dr. Sherry, we see that she does teach courses,
[06:59.420 --> 07:05.180]  like hence the book tour. She sells books. She trains folks. She has about 50, 60,000 followers
[07:05.180 --> 07:11.400]  on Twitter, not shown. She also has a big presence on Facebook and also on newsletters. So stuff that
[07:11.400 --> 07:15.520]  you don't even see. And she gets paid to do it. Now it's kind of like her
[07:16.220 --> 07:20.240]  beliefs and her livelihood are the same thing. So she's very invested in this.
[07:22.660 --> 07:28.420]  But once we start looking at it, there are kind of a few things that looks a little strange.
[07:28.740 --> 07:36.180]  So she talks about Nazis in her profile, like that's the first thing, that's the first word
[07:36.180 --> 07:42.520]  she shows you. And so that kind of is kind of like a whistle to one side of the house.
[07:42.660 --> 07:47.160]  And then talks about abuses by the state. So that's kind of like a different kind of thing.
[07:47.160 --> 07:52.000]  And so it's kind of interesting to ask, like, who is the community around her?
[07:52.180 --> 08:03.220]  So let's take a look. So what I did here is loaded in busy Dr. T. And so we're looking at her,
[08:03.220 --> 08:09.040]  and what we see in colors around her are basically, she's only following a couple hundred
[08:09.040 --> 08:13.840]  people, but she has 50,000 people following her. And so when we look at a sample of that,
[08:13.840 --> 08:19.980]  what we see up top is a bunch of folks, essentially public, basically anti-vax,
[08:20.980 --> 08:24.960]  natural medicine, a lot of kind of medically-minded folks that you'd sort of expect
[08:24.960 --> 08:30.440]  to be here. But then the surprise is once we start seeing the other colors.
[08:30.440 --> 08:36.320]  And so the tools here just kind of automatically colored and laid it out and everything,
[08:36.320 --> 08:40.800]  like it's kind of just traditional graph analytics. So we see Donald Trump,
[08:40.800 --> 08:47.620]  we see Michelle Malkin, we see Candace. Basically, like in the dark green,
[08:47.620 --> 08:51.800]  we're seeing these public figures, kind of national conservative public figures.
[08:52.640 --> 08:58.720]  On the right, we're seeing, but a bit on the right, and then actually, especially down here,
[08:58.720 --> 09:04.360]  we're seeing a mixture of QAnon, which is a conspiracy group we're hearing about a lot
[09:04.360 --> 09:11.380]  nowadays. We're seeing a lot of MAGA and Trump supporters. So not super surprising given we
[09:11.380 --> 09:16.820]  had Trump. And so it's kind of interesting that we have on the top, we have like, who do you expect
[09:16.820 --> 09:22.340]  with anti-vax? But on the bottom, we have like the whole MAGA thing. We have conspiracy theorists
[09:22.340 --> 09:27.660]  from QAnon. They're actually even like UFO stuff going on, on the bottom. It's kind of like
[09:27.660 --> 09:32.840]  interesting to see that mix. So I want to point out another phenomenon here.
[09:33.260 --> 09:37.980]  When we're looking at each community, like the question is like, how are they interacting? Is it
[09:37.980 --> 09:43.480]  like they all really love our Dr. Busy T? Like she just wrote this wonderful book. What we actually
[09:43.480 --> 09:49.720]  see is like, for example, from the QAnon zone, they're reaching into the medical zone. So they're
[09:49.720 --> 09:55.560]  following folks there. Conversely, if we pick somebody big in the kind of this top region,
[09:55.560 --> 10:00.420]  sometimes it's localized, but other times you see it's reaching into the QAnon zone.
[10:00.820 --> 10:05.560]  Kind of no matter which cluster I do, you're going to see those reach outs.
[10:07.120 --> 10:15.040]  Excuse me. So that's actually kind of strange, where there's no reason to believe one conspiracy
[10:15.460 --> 10:20.640]  has anything to do with another. And so we're actually kind of almost like in this other world.
[10:21.520 --> 10:28.520]  And what we're seeing is the networks do cross over. And so the first thing is what we're seeing
[10:28.520 --> 10:35.700]  is the question is why? And that's basically to support influence. Sometimes that could be direct
[10:35.700 --> 10:39.520]  where you're just retweeting from each other's. Other times it might be to trick the Twitter
[10:39.520 --> 10:44.460]  algorithm. So for example, here is actually a Fox News reporter. So it's someone on TV.
[10:45.560 --> 10:52.660]  They have 40,000 followers, which is reasonable for a personality, but they're following 30,000
[10:52.660 --> 10:57.620]  people back. So that probably suggests an inauthentic and automatic action of following
[10:57.620 --> 11:02.200]  back somehow. We don't know what order they happened in, but in general, that's prohibited
[11:02.200 --> 11:09.040]  by Twitter's terms of service. But they get clicks. And so even though it violates terms
[11:09.040 --> 11:15.940]  of service, there they are. And that's basically the welcome to the misinformation network.
[11:17.360 --> 11:22.920]  And I realized that I glossed over this, but the reason we care about Dr. T and that she's
[11:22.920 --> 11:29.820]  talking about this stuff is that she and her community are already pushing messages against
[11:30.040 --> 11:35.520]  a COVID vaccine, even though one hasn't even passed trials yet. She's against it before it's
[11:35.520 --> 11:41.100]  even been tested, just by principle. And so Twitter's been supporting her in doing that,
[11:41.100 --> 11:47.440]  essentially. And I'll get to that in a second. So I do want to take one more, I want to go a
[11:47.440 --> 11:54.120]  little bit deeper into our Dr. T. And so in particular, I want to understand the communities,
[11:54.120 --> 11:59.300]  like what they're doing. So here what we're going to do, and I want to see if there's kind of extra
[11:59.300 --> 12:02.900]  coordination structure. And so we're getting a bit more on the data science side. So what's
[12:02.900 --> 12:08.340]  going on here is we took the accounts, we actually took a look at about 100 million COVID-related
[12:08.340 --> 12:14.120]  tweets to see if there's kind of organized behavior within it. For each account, we just
[12:14.120 --> 12:20.080]  tried to look at for known sort of misinformation topics, what's sort of the signature, like, you
[12:20.080 --> 12:25.380]  know, how many times did it like activate on different topics. And for each account, we now
[12:25.380 --> 12:30.640]  have that sort of like misinformation fingerprint, essentially. And then we ran something called
[12:30.640 --> 12:37.040]  UMAP. I think it's a universal manifold approximation or something like that. But basically
[12:37.040 --> 12:44.100]  tries to smooth out and assume structure behind that. And then we took a look. So let's try it
[12:44.100 --> 12:51.820]  out. So yeah, so what's going on here is we're taking a look at those accounts. And we only
[12:51.820 --> 12:58.080]  are showing from all the accounts, we're looking at the 5000 or so that we actually saw active
[12:58.080 --> 13:04.160]  across multiple misinformation topics. And so these are kind of the really the ones that have
[13:04.160 --> 13:08.960]  kind of that behavior going across topics like are just like what's going on.
[13:11.020 --> 13:16.320]  And we ran UMAP. But then what we also did, you're seeing here, I'm drawing these little edges.
[13:16.320 --> 13:22.660]  We also ran a afterwards k-nearest neighbor and put them on screen. So that whenever we see
[13:22.660 --> 13:29.500]  accounts, kind of tweeting on similar topics and similar intensities. So it's really, for example,
[13:29.500 --> 13:33.860]  imagine two bots program to kind of talk about the same thing, we're controlled by the same person to
[13:33.860 --> 13:40.160]  kind of, you know, retweet someone else. That's when we're going to draw that edge. And so again,
[13:40.160 --> 13:44.100]  they're not necessarily following, they're actually just on certain topics, they're actually pushing
[13:44.100 --> 13:49.960]  those topics in the same way. So it's entirely content based. And what we're seeing here is
[13:49.960 --> 13:56.240]  some interesting structure pops out for across COVID Twitter. And then the second thing I did
[13:56.240 --> 14:05.800]  here is coloring, you're seeing some of these are a bit pinkish. So if we jump, let's take a look at
[14:05.800 --> 14:12.180]  this one. So for this kind of group of activity, we're seeing a lot of this q&a and stuff. And so
[14:12.180 --> 14:19.420]  what I'm doing here is if in those tweets talking about COVID, they also talk about q&a. And we have
[14:19.520 --> 14:24.760]  a couple words on that, we call that out too. And so what we're seeing is like, you know, for this
[14:24.760 --> 14:31.380]  general topic, a bunch of accounts never explicitly mentioned q&a. But it's definitely maybe it's like
[14:31.380 --> 14:36.700]  two coordinators, or who knows, or like something like that. But we're actually seeing a lot of q&a
[14:36.700 --> 14:43.260]  influence across these different organized groups. And actually, in particular, almost all of these
[14:43.260 --> 14:48.520]  have a bit of q&a. So you're seeing it's not all white, it's all a little off white. And then what's
[14:48.520 --> 14:53.520]  more, you're seeing some of them are you're seeing these pockets of heaviness. And we calculated
[14:53.520 --> 15:00.240]  out about 15% of these accounts spread throughout our q&a. And so at least for COVID Twitter,
[15:00.240 --> 15:03.520]  that was actually kind of interesting. And for us, that was a surprise.
[15:03.800 --> 15:10.600]  So that kind of put q&a on in terms of medical misinformation on our radar. So that's not good.
[15:12.920 --> 15:25.360]  Okay, so jumping ahead. So okay, so we're looking at our busy Dr. T. And clearly blatant activity
[15:25.360 --> 15:30.840]  going on here. The last thing I want to point out was, then we look on the right here, and Twitter
[15:30.840 --> 15:35.520]  is saying, hey, you should probably now that you're here, you should also check out this
[15:37.340 --> 15:44.320]  Judy Mikovits. This PhD sounds authentic, right? Sounds smart. You may have also known about this
[15:44.920 --> 15:52.140]  Judy as a video going around called Plandemic, which, you know, got pulled down as pretty bad
[15:52.140 --> 15:57.220]  misinformation. And so that was pretty shocking to see, like, I was like, oh, that name sounds
[15:57.220 --> 16:01.460]  familiar. So that was a little crazy. That was like the number one recommendation of Twitter
[16:01.460 --> 16:06.540]  as soon as you go to this web page is to learn about Plandemic. So and this is like, you know,
[16:06.540 --> 16:13.220]  way after the fact. And so kind of in summary, what we're seeing in this, in this basic, pretty
[16:13.220 --> 16:19.980]  basic, honestly, analysis, very blatant anti-vaccine misinformation, a lot of it about
[16:19.980 --> 16:25.360]  COVID in particular, we're seeing it's not just anti-vaxxers, but we're actually seeing a lot of
[16:25.360 --> 16:32.200]  mis, cross-community support by misinformers. It's pretty blatant. It's also it's pretty,
[16:32.200 --> 16:37.500]  pretty obvious how they abuse, like, the very few terms of service Twitter does have.
[16:37.500 --> 16:46.780]  So that's a little strange. It is a little disappointing to see Fox, the Republican Party,
[16:46.780 --> 16:53.500]  MAGA, and a lot of them kind of involved in this anti-medical misinformation. And again,
[16:53.500 --> 17:00.740]  it's so blatant. And then on the platform side, it's when we saw that overlap and like,
[17:00.740 --> 17:03.680]  oh, wait a minute, Twitter makes money off of some of this stuff.
[17:04.740 --> 17:09.540]  They're clearly looking the other way for the terms of service violation. And then like,
[17:09.540 --> 17:14.480]  what was the the the Plandemic promotion? Just that was just that blew my mind. So
[17:15.080 --> 17:21.140]  this is not, this is not a healthy situation. So from a data perspective,
[17:23.500 --> 17:27.480]  or actually security analytics perspective, I do want to step back.
[17:27.960 --> 17:34.180]  If we think about attackers, defenders, ultimately, the targets are individuals and organizations
[17:34.180 --> 17:40.420]  going on a radicalization journey. And so every time Twitter recommends you do another crazy thing
[17:40.420 --> 17:48.120]  that just takes you deep, deeper. Interesting extra level on the threat model here is that
[17:48.120 --> 17:54.380]  folks are essentially influenced based on their life conditions. So maybe they're going through a
[17:54.380 --> 17:58.740]  hard time, maybe they're elderly and don't really know what's going on. And also by their community,
[17:58.740 --> 18:03.100]  you know, if your if your pastor or somebody or your family friends tells you something,
[18:03.100 --> 18:09.020]  you're more inclined to believe it. And so being able to infect communities with this stuff is
[18:09.020 --> 18:12.060]  having mastery over that and platform level support, it's not good.
[18:12.680 --> 18:19.880]  For actors, it's a lot of what you'd expect. I think that the surprising thing for for threat
[18:19.880 --> 18:26.820]  actors here is we actually consider the, at this point, the, the platforms to be part of the
[18:26.820 --> 18:34.280]  threat. In particular, we we break down the platforms as the social network system. So
[18:34.860 --> 18:38.700]  Facebook owns a bunch of social network, Google owns a bunch of social networks,
[18:38.880 --> 18:42.800]  a few are just big on their own, right, like Reddit, Reddit and Twitter. And, you know,
[18:42.800 --> 18:48.540]  Twitter is a multi billion dollar profitable company. But here we are. A little less obvious
[18:48.540 --> 18:54.040]  to folks is messaging. And again, actually, a lot of the same companies like Google, Microsoft,
[18:54.040 --> 19:00.700]  Yahoo, own essentially the messaging channels, those are completely unmonitored. And so a lot
[19:00.700 --> 19:05.800]  of times the radicalization actually hops over to newsletters. So that's stuff basically dark data.
[19:06.680 --> 19:15.360]  And also the local news, I think like Fox and folks like that. A little surprising is there's
[19:15.360 --> 19:21.400]  kind of a monopoly effect where because you do have regional different ways of having monopolies
[19:21.400 --> 19:26.840]  on people's attention. That's all the reason these companies got so big. Same reason we're
[19:26.840 --> 19:30.980]  basically at the whimsy of these companies. And so when you kind of calculate it out,
[19:30.980 --> 19:39.280]  the 80% essentially is just about 13 of these groups. And so if we have a fickle Silicon Valley
[19:39.280 --> 19:44.860]  CEO, if we have employees who just go along for the ride, if we have advertisers keep feeding
[19:44.860 --> 19:54.160]  them money, you know, so that's kind of from the threat model perspective. Shifting gears a bit to
[19:54.160 --> 20:00.220]  more like kind of data like security, data engineering, data science. In order to do the
[20:00.220 --> 20:05.420]  data science, I did want to talk a little bit about this. We end up having to build a lot to
[20:05.420 --> 20:09.620]  work at kind of the speeds and scales we're targeting. So if you want to have an instant
[20:09.620 --> 20:13.780]  response pipeline, that means you have to kind of have streaming, you know, can you take hundreds
[20:13.780 --> 20:17.820]  or thousands of things a second. So our V1, you're kind of seeing the performance of that
[20:17.820 --> 20:22.540]  where you're architecting to a V2 so that we can actually go to arbitrary scale, just have
[20:22.540 --> 20:28.620]  essentially budget limited. You are noticing we're doing a lot of graph technology. And so basically
[20:28.620 --> 20:32.860]  that's a nice way to deal with a lot of heterogeneous data and a lot of like linkage of
[20:32.860 --> 20:37.280]  data, even actually in a machine learning world where instead of just direct links, you might have
[20:37.280 --> 20:42.400]  vectors, feature vectors. Two other interesting things about the stack I did want to point out
[20:42.400 --> 20:48.060]  here. One is we do do end-to-end GPU computing just to be able to, you know, you can actually,
[20:48.060 --> 20:51.540]  if you want to have a billion node graph in memory, you actually could do that stuff with,
[20:51.540 --> 20:57.560]  for example, a Rapids AI stack. Likewise, neural nets in this space, graph neural nets are getting
[20:57.560 --> 21:02.800]  interesting. We actually do more today on the traditional ones, but we're looking at those.
[21:02.940 --> 21:06.160]  And the last part, and I think most important, is kind of division of team.
[21:06.580 --> 21:12.300]  We have basically heroes for data engineers. And then a big part of this is figuring out ways to
[21:12.300 --> 21:16.800]  get out of the way, just given it is an open volunteer effort. And so we actually do a lot
[21:16.800 --> 21:22.480]  on the self-service. And that's a big theme for look about powering. So think notebooks,
[21:22.480 --> 21:29.120]  streamlet dashboards, graphistry, do visual analytics and automatic ML.
[21:29.120 --> 21:34.680]  Just increasing the ability of people just to go and handle the stuff.
[21:37.040 --> 21:45.480]  From the data science side, I kind of wanted a fun example here. For example, we had a
[21:45.480 --> 21:50.740]  news organization ask us about Obamagate. And so one of our analysts, Cody, did a good job of using
[21:51.270 --> 21:58.480]  hashtag analysis to connect the dots to a month back to some QAnon stuff. And it was kind of
[21:58.480 --> 22:04.620]  impressive because the Trump tweet made no sense at all, even when I read it today. All these kind
[22:04.620 --> 22:14.100]  of code words. But once we moved to more modern NLP techniques and go away from exact match of
[22:14.100 --> 22:22.340]  terms. So Cody had to look for Obamagate. But with NLP, we can do things like look for the word
[22:22.340 --> 22:28.780]  Obama and the word conspiracy. But not the word, but just the fuzzy topic. And when you have those
[22:28.780 --> 22:36.660]  things, then we can kind of more loosely kind of walk back. And so that means we can do kind of
[22:36.660 --> 22:43.880]  traditional analysis more easily. And so we've been building out a lot around that. I want to
[22:43.880 --> 22:50.980]  modeling from basically what everyone does now. And then the other thing I want to point out here
[22:50.980 --> 22:55.060]  is this is also opening up types of analysis that just were straight up, like, you couldn't even
[22:55.060 --> 23:03.140]  think about. So maybe one example could be for looking at masks in a community where you might
[23:03.140 --> 23:09.360]  ask, like, which messages are against mask use? That's kind of a strange... that's kind of a
[23:09.360 --> 23:15.160]  gets into semantics and meaning a bit. And that's hard to do just by keywords. And then
[23:15.160 --> 23:18.900]  could you summarize it? Could we actually have a demographic breakdown? There are just certain
[23:18.900 --> 23:23.840]  things that with traditional NLP, you couldn't have done. But over the years, over the last,
[23:23.840 --> 23:28.640]  like, five years, and actually even two years, it's just been really nuts what's been going on.
[23:33.160 --> 23:36.840]  Then kind of the last piece I'm going to share here is kind of just where we are,
[23:36.840 --> 23:43.120]  where we're going. You're kind of getting a sense that we can now bring in a researcher on a topic,
[23:43.120 --> 23:48.200]  and if we have the data, or we help them pull in the data, and kind of do this kind of analysis.
[23:48.520 --> 23:51.860]  Next level of empowerment we're looking at is, like, as we're building this out,
[23:51.860 --> 23:55.780]  is can we actually partner with less technical organizations, and if we have to kind of do this
[23:55.780 --> 24:00.160]  to protect their community? Imagine, like, a local civic leader talking about the
[24:00.160 --> 24:05.700]  cancer community with Andrea, with kind of at-risk patients. And there we're looking at,
[24:05.700 --> 24:12.340]  are there folks who are not even maliciously misinforming in their community, but people
[24:12.340 --> 24:16.280]  are just confused? Or are there people kind of targeting them? Can we start mapping out and
[24:16.280 --> 24:22.240]  building tools and stuff around that? So that's kind of the phase shift we're starting to go into,
[24:22.240 --> 24:26.880]  so that's pretty exciting. You're kind of seeing, as you build out that state of capability,
[24:26.880 --> 24:33.180]  all this stuff unlocks. And then, finally, I just kind of want to close out and tie together
[24:33.180 --> 24:37.400]  some of the stuff here. So the beginning, I was kind of sharing that investigation.
[24:38.300 --> 24:41.980]  Hopefully, what you got across is, like, yeah, the analysis is cool, and we're able to, like,
[24:42.380 --> 24:48.440]  do these giant analyses. But, like, ultimately, a lot of the important stuff is so in-your-face
[24:48.440 --> 24:55.160]  and so huge numbers that, like, again, that analogy of fishing with dynamite. And it's,
[24:55.160 --> 25:02.200]  you know, those 13 companies, they shouldn't be gifting, essentially, millions or billions of
[25:02.200 --> 25:09.000]  dollars in free targeted ads to these groups and targeted recommendations. It's kind of that stuff
[25:09.000 --> 25:16.680]  that I think is more systemic. I am disappointed to see kind of these otherwise legitimate groups
[25:16.680 --> 25:22.000]  appear to be corrupted by misinformation. So call out, essentially, we're seeing MAGA is just so
[25:22.000 --> 25:26.740]  ever-present here. We're seeing social media companies, regular media companies. So there
[25:26.740 --> 25:31.780]  are things you could do if you're an employee, if you are anywhere near the advertising spend of
[25:31.780 --> 25:38.100]  your company. All that stuff's important. And then kind of turning it around,
[25:39.180 --> 25:45.760]  that's all kind of social policy. A lot of folks tuning in are technologists of different stripe.
[25:46.940 --> 25:54.440]  And the bad news is we're seeing these monopolistic media companies are not able to
[25:54.440 --> 25:58.280]  self-regulate. Like, the left hand and the right hand are always at each other, and it's kind of
[25:58.280 --> 26:05.380]  brought us to where we are. Likewise, they've kind of gutted investigative and journalist teams
[26:06.000 --> 26:10.120]  just because their ad companies sucked all the money. And likewise, we're seeing private
[26:10.120 --> 26:17.020]  companies are sort of compromised as they're kind of trying to deal with revenue. So that's bad.
[26:17.160 --> 26:23.320]  What's good is we actually have an open source community. We, I mean, just like the technology
[26:23.320 --> 26:29.140]  community, that is huge. And now at the same time, data science is, you know, our Python that's almost
[26:29.140 --> 26:38.480]  like built into the DNA. So I see strong potential and gathering potential for that. I encourage that.
[26:38.880 --> 26:45.340]  You're seeing some of that happen here with Project Domino. And so on that point, if you are
[26:45.340 --> 26:51.380]  into this kind of stuff and you could help out with coding, head into our Slack. Likewise, as we
[26:51.380 --> 26:58.400]  launch in the coming months, these kind of more self-serve and tiers for folks sitting on top,
[26:58.400 --> 27:05.300]  reach out for partnering. And likewise, while we are looking at doing sustainable approaches to our
[27:05.300 --> 27:12.520]  growth, especially as sort of a bootstrap phase that we're in for the next year or so,
[27:12.520 --> 27:16.100]  we're definitely, you know, grants, whatever, that stuff does help with the core team
[27:16.100 --> 27:20.500]  do what they need to do. Likewise, you know, serverless compute time, anything like that.
[27:21.380 --> 27:27.680]  With that, thank you for your time. And hopefully this changed a little bit about either how
[27:27.680 --> 27:35.020]  misinformation is working in practice in the COVID era. And also, even more importantly,
[27:35.020 --> 27:39.900]  that we actually could do things and we actually detect it and then can actually build tools for it.
[27:40.280 --> 27:41.280]  Thank you.