1 00:00:00,160 --> 00:00:02,320 Speaker 1: Some insight for you now on how the tech giants 2 00:00:02,360 --> 00:00:05,760 Speaker 1: are handling what is an increasingly hostile view of social media. Australia, 3 00:00:05,800 --> 00:00:07,880 Speaker 1: of course, has seen millions of kids account shut down 4 00:00:07,920 --> 00:00:10,400 Speaker 1: as a result of their ban. Are the UK potentially 5 00:00:10,440 --> 00:00:12,920 Speaker 1: going down that same route. France has already put legislation 6 00:00:13,000 --> 00:00:16,200 Speaker 1: into their parliament. One of matters most senior executives happens 7 00:00:16,200 --> 00:00:17,520 Speaker 1: to be in the country for what they call a 8 00:00:17,560 --> 00:00:20,599 Speaker 1: safety camp and Tikeny Davis is the VPN global head 9 00:00:20,640 --> 00:00:23,200 Speaker 1: of safety at Meta and is with us. Very good 10 00:00:23,200 --> 00:00:27,200 Speaker 1: morning to you. Just for context, explain where in the 11 00:00:27,240 --> 00:00:29,560 Speaker 1: line or chain of things you are to Mark Zuckerberg, 12 00:00:29,840 --> 00:00:31,040 Speaker 1: how important are you? 13 00:00:31,320 --> 00:00:33,760 Speaker 2: I feel like your description was sort of the best. 14 00:00:34,520 --> 00:00:37,160 Speaker 2: I work on teams safety at METTA. I've been here 15 00:00:37,200 --> 00:00:40,239 Speaker 2: for eleven years. I have a team that works with 16 00:00:40,520 --> 00:00:43,599 Speaker 2: all of the teams across the company to come up 17 00:00:43,640 --> 00:00:46,560 Speaker 2: with our strategy. So I'm right in the center of it. 18 00:00:46,880 --> 00:00:49,320 Speaker 1: Do you know or could you isolate out in any 19 00:00:49,360 --> 00:00:52,559 Speaker 1: given day as part of your job, which strand that 20 00:00:52,600 --> 00:00:54,800 Speaker 1: Meta has. So if you look at Facebook, inst what's 21 00:00:54,840 --> 00:00:59,080 Speaker 1: apt Messenger and threads, where's the trouble safety wise in 22 00:00:59,120 --> 00:00:59,920 Speaker 1: your organization? 23 00:01:00,640 --> 00:01:02,840 Speaker 2: I don't know if I would say there's trouble in 24 00:01:02,920 --> 00:01:05,360 Speaker 2: any one app more than another, But I do think 25 00:01:05,400 --> 00:01:09,600 Speaker 2: people are paying close attention to Instagram because so many teens. 26 00:01:09,360 --> 00:01:11,840 Speaker 1: Are on it is Instagram a problem? 27 00:01:12,280 --> 00:01:14,120 Speaker 2: I don't think of Instagram as a problem. I think 28 00:01:14,160 --> 00:01:17,360 Speaker 2: most teens see Instagram as a really effective way to 29 00:01:17,400 --> 00:01:21,360 Speaker 2: build community, to engage in their interests, etc. But I 30 00:01:21,400 --> 00:01:24,559 Speaker 2: do understand parents, and I do understand their concerns about 31 00:01:24,560 --> 00:01:27,399 Speaker 2: trying to provide a safe environment. I am a parent 32 00:01:27,440 --> 00:01:29,759 Speaker 2: I have she's twenty seven now, and I worried about 33 00:01:29,760 --> 00:01:31,880 Speaker 2: those things. So I do understand why parents are worried 34 00:01:31,880 --> 00:01:32,240 Speaker 2: about it. 35 00:01:32,240 --> 00:01:35,560 Speaker 1: For sure, what's the most agy safety issue going on? 36 00:01:35,640 --> 00:01:37,160 Speaker 1: As far as you're concern. 37 00:01:37,480 --> 00:01:40,040 Speaker 2: Edgy, the most edgy safety issue I think you worry 38 00:01:40,040 --> 00:01:43,560 Speaker 2: about most. I think probably the thing that I worry 39 00:01:43,600 --> 00:01:47,480 Speaker 2: about most is how do we make it simple for parents? Candidly, 40 00:01:47,520 --> 00:01:49,800 Speaker 2: for parents, I think it's been very hard for them. 41 00:01:50,080 --> 00:01:53,720 Speaker 2: It either involves grabbing that phone from teens, and so 42 00:01:53,880 --> 00:01:55,560 Speaker 2: what I'd like to see or what we'd like to 43 00:01:55,560 --> 00:01:57,080 Speaker 2: see is a way for parents to be able to 44 00:01:57,120 --> 00:01:59,120 Speaker 2: sort of say yes or no and approve the download 45 00:01:59,160 --> 00:01:59,520 Speaker 2: of an app. 46 00:01:59,640 --> 00:02:02,120 Speaker 1: Okay, you'd have to have parents involved. Blah, what do 47 00:02:02,200 --> 00:02:04,280 Speaker 1: you Mike, all of Australian world I've done. 48 00:02:05,920 --> 00:02:08,480 Speaker 2: I get it. I understand why they're concerned about it. 49 00:02:08,639 --> 00:02:11,120 Speaker 2: I think the impetus to go to a band is 50 00:02:11,160 --> 00:02:14,600 Speaker 2: that it provides a very neat and simple solution. But 51 00:02:14,760 --> 00:02:17,920 Speaker 2: I think it's probably a little bit of a fool's 52 00:02:17,960 --> 00:02:20,079 Speaker 2: air and in the sense that teens are going to 53 00:02:20,280 --> 00:02:23,639 Speaker 2: move outside and around bands. You can't make a band 54 00:02:23,720 --> 00:02:26,040 Speaker 2: for the entire internet. So how do you find a 55 00:02:26,160 --> 00:02:28,760 Speaker 2: way to really put in place what they're trying to do, 56 00:02:28,840 --> 00:02:31,480 Speaker 2: which is to basically create friction? And this is where 57 00:02:31,520 --> 00:02:34,679 Speaker 2: I think if you put parental approval, So anytime you 58 00:02:34,760 --> 00:02:37,480 Speaker 2: download an app, a team goes to download an app, 59 00:02:37,600 --> 00:02:39,839 Speaker 2: a parent has to say yes or no. In other words, 60 00:02:39,880 --> 00:02:43,680 Speaker 2: you're starting with a restriction around entering and you're basically 61 00:02:43,800 --> 00:02:46,280 Speaker 2: saying Okay, yes you can do this or no you can't. 62 00:02:46,720 --> 00:02:49,639 Speaker 2: That's a better way to do that across the entire Internet. 63 00:02:49,880 --> 00:02:53,000 Speaker 1: So how come you could nague that successfully to the Australians. 64 00:02:53,200 --> 00:02:56,040 Speaker 2: I think Australia moved very quickly. They didn't have a 65 00:02:56,160 --> 00:02:58,400 Speaker 2: chance to really consult with experts that came up during 66 00:02:58,440 --> 00:03:00,840 Speaker 2: the time. I think if we'd had more time, we 67 00:03:00,880 --> 00:03:03,000 Speaker 2: would have been in a better position to make that case. 68 00:03:03,280 --> 00:03:05,320 Speaker 1: Are you hoping, and I don't want to make this 69 00:03:05,440 --> 00:03:08,639 Speaker 1: overtly political, but are you hoping it doesn't work? And 70 00:03:08,840 --> 00:03:11,160 Speaker 1: in not working, you can then go to the next government, 71 00:03:11,560 --> 00:03:13,440 Speaker 1: off the cab, the cab, off the rank to say 72 00:03:13,520 --> 00:03:13,880 Speaker 1: there it is. 73 00:03:14,160 --> 00:03:16,720 Speaker 2: Yeah, it's a good question. I would never hope for 74 00:03:16,840 --> 00:03:20,160 Speaker 2: something not to work that's designed to make things better 75 00:03:20,520 --> 00:03:22,799 Speaker 2: for families. What I'd really hope for is that we 76 00:03:22,880 --> 00:03:25,160 Speaker 2: can develop a solution that does. 77 00:03:25,080 --> 00:03:27,840 Speaker 1: Work for the Do you have a solution in the 78 00:03:27,960 --> 00:03:32,240 Speaker 1: dark corners of meta that could technically achieve what they 79 00:03:32,360 --> 00:03:34,440 Speaker 1: want to do, but you just don't want to do 80 00:03:34,560 --> 00:03:37,040 Speaker 1: it because this is all about you losing customer buz it? 81 00:03:37,200 --> 00:03:39,960 Speaker 2: Yeah, No, I think it's it's actually really really hard 82 00:03:40,040 --> 00:03:43,720 Speaker 2: to do what they are asking us to do. There's 83 00:03:43,800 --> 00:03:47,720 Speaker 2: no simple way without asking everybody to provide some kind 84 00:03:47,760 --> 00:03:52,119 Speaker 2: of an ID to verify, to verify age that means 85 00:03:52,160 --> 00:03:55,120 Speaker 2: that every app and kids use about teens is about 86 00:03:55,160 --> 00:03:58,400 Speaker 2: forty apps and then taking that information. This is again 87 00:03:58,480 --> 00:04:00,520 Speaker 2: why we're going to this os app or approach. 88 00:04:00,800 --> 00:04:03,360 Speaker 1: So the other thing is in Australia this last week, 89 00:04:03,800 --> 00:04:06,840 Speaker 1: there was a story floating around. So the daughter this 90 00:04:07,040 --> 00:04:09,880 Speaker 1: is for it was for Instant or Watts, I can't remember, 91 00:04:11,320 --> 00:04:14,680 Speaker 1: puts the phone to the dad's face and the company 92 00:04:14,800 --> 00:04:18,120 Speaker 1: that you guys employed don't do age. They do face 93 00:04:18,240 --> 00:04:21,040 Speaker 1: or they just do face not age whatever. Anyway, the 94 00:04:21,120 --> 00:04:23,320 Speaker 1: point is she got back on her account, right. I 95 00:04:23,400 --> 00:04:26,520 Speaker 1: don't know particular, but in general as a thing. 96 00:04:26,680 --> 00:04:29,120 Speaker 2: Yeah, in general. I mean one of the challenges with 97 00:04:29,240 --> 00:04:32,720 Speaker 2: all these different technologies is that they work well, but 98 00:04:32,839 --> 00:04:34,680 Speaker 2: there are ways to get around them, and a lot 99 00:04:34,760 --> 00:04:38,000 Speaker 2: of you know, people spend time trying to get around them. 100 00:04:38,320 --> 00:04:41,159 Speaker 2: We've done sort of a multi layered approach to try 101 00:04:41,240 --> 00:04:44,520 Speaker 2: to try to address that. But there is this moment 102 00:04:44,560 --> 00:04:47,799 Speaker 2: in time where teens and parents basically purchase the phone, 103 00:04:48,040 --> 00:04:50,920 Speaker 2: and parents have an opportunity to be set the age. 104 00:04:51,320 --> 00:04:55,120 Speaker 2: And if we use that moment in time, well, we'd 105 00:04:55,160 --> 00:04:57,320 Speaker 2: be in a much better process to create the right 106 00:04:57,400 --> 00:04:58,799 Speaker 2: sort of multi layered approach. 107 00:04:59,080 --> 00:05:01,240 Speaker 1: Was this always coming for your sort of companies? This 108 00:05:01,720 --> 00:05:03,640 Speaker 1: as you put at this moment in time, you know, 109 00:05:04,200 --> 00:05:06,920 Speaker 1: teen years ago, when this is all like new and fantastic, 110 00:05:07,000 --> 00:05:09,800 Speaker 1: did somebody in the dark corners of Meta. Go tell 111 00:05:09,839 --> 00:05:11,680 Speaker 1: you what, someone at some point is going to push 112 00:05:11,760 --> 00:05:14,040 Speaker 1: back big time and we need to prepare for this. 113 00:05:14,440 --> 00:05:17,160 Speaker 2: Well, I've I've been at Meta for about eleven years, 114 00:05:17,200 --> 00:05:19,479 Speaker 2: so I do think over the course of the entire 115 00:05:19,600 --> 00:05:21,400 Speaker 2: time that I've been there, we have been thinking about 116 00:05:21,680 --> 00:05:24,880 Speaker 2: how do we make our app safe for young people? 117 00:05:25,040 --> 00:05:27,359 Speaker 2: So we have teen accounts that put kids into built 118 00:05:27,400 --> 00:05:30,719 Speaker 2: in restrictions and only they can only change them if 119 00:05:30,720 --> 00:05:34,240 Speaker 2: you're under sixteen, if their parents give them permission. That's 120 00:05:34,320 --> 00:05:36,760 Speaker 2: all a long journey over ten years of putting in 121 00:05:36,800 --> 00:05:39,880 Speaker 2: place over fifty different tools, and throughout this process we've 122 00:05:39,960 --> 00:05:42,200 Speaker 2: been thinking about how to make it simple. When I 123 00:05:42,320 --> 00:05:45,479 Speaker 2: came in eleven years ago, my daughter who's now twenty seven, 124 00:05:45,880 --> 00:05:48,279 Speaker 2: was right in the thick of it, and I had 125 00:05:48,360 --> 00:05:51,520 Speaker 2: this urgency myself of how do we make something simpler 126 00:05:51,560 --> 00:05:51,760 Speaker 2: for me? 127 00:05:52,360 --> 00:05:56,840 Speaker 1: How much of your thinking at Meta is about the 128 00:05:57,000 --> 00:06:00,680 Speaker 1: idea that really, you gotta have the pairs on board, 129 00:06:00,760 --> 00:06:04,000 Speaker 1: all the kids on board. And the alternative to that is, 130 00:06:04,160 --> 00:06:07,800 Speaker 1: let's all blame big tech because they're rich and successful 131 00:06:07,839 --> 00:06:11,960 Speaker 1: and they don't care. So that interface between those two conversations, Yeah, 132 00:06:12,279 --> 00:06:13,479 Speaker 1: you know, look, we all have. 133 00:06:13,800 --> 00:06:16,040 Speaker 2: A role to play. If you think about anything with 134 00:06:16,279 --> 00:06:19,600 Speaker 2: young people, we basically say, okay, what are the rules 135 00:06:19,600 --> 00:06:21,400 Speaker 2: that we want to have in place as a society. 136 00:06:21,400 --> 00:06:24,400 Speaker 2: You're on a platform for ours are the what are 137 00:06:24,440 --> 00:06:28,240 Speaker 2: the obligations that we have as a company or you know, government, 138 00:06:28,400 --> 00:06:30,640 Speaker 2: what are the obligations that we have to create certain 139 00:06:30,720 --> 00:06:33,800 Speaker 2: safety safety mechanisms? And then we always see it as 140 00:06:33,839 --> 00:06:37,240 Speaker 2: a partnership with parents, because parents know a specific thing 141 00:06:37,279 --> 00:06:39,400 Speaker 2: about their child that you or I might not know. 142 00:06:39,800 --> 00:06:42,479 Speaker 2: And I say parents, but I really should say caregivers 143 00:06:42,520 --> 00:06:43,200 Speaker 2: because not everyone. 144 00:06:43,279 --> 00:06:45,360 Speaker 1: Of course, were you? You didn't happen to be You 145 00:06:45,360 --> 00:06:48,640 Speaker 1: remember when Zucker bog fronted up at the congressional hearing. 146 00:06:48,720 --> 00:06:49,960 Speaker 1: It was on the hill. Do you remember that when 147 00:06:50,160 --> 00:06:52,520 Speaker 1: it was yeah, yeah, but the last one where he 148 00:06:52,600 --> 00:06:54,880 Speaker 1: had to turn around to the families who were standing right, Okay, 149 00:06:54,880 --> 00:06:56,920 Speaker 1: were you there? You didn't happen to be there. I 150 00:06:57,080 --> 00:06:58,800 Speaker 1: wasn't in the realm Okay, but you will have seen 151 00:06:58,839 --> 00:07:00,920 Speaker 1: that obviously. How do you handle it? What do you 152 00:07:00,960 --> 00:07:03,120 Speaker 1: say to those families who look at you and go, 153 00:07:03,360 --> 00:07:04,480 Speaker 1: it's it's on you. 154 00:07:04,880 --> 00:07:08,400 Speaker 2: Yeah, you know, here's the thing. It's a terrible tragedy 155 00:07:09,000 --> 00:07:12,720 Speaker 2: and you can't. You can't dismiss those tragedies. It's you know, 156 00:07:13,040 --> 00:07:17,280 Speaker 2: for those parents, they're dealing with something that's extraordinarily horrible 157 00:07:17,400 --> 00:07:20,120 Speaker 2: and that they should never have had had to deal with. 158 00:07:20,760 --> 00:07:24,160 Speaker 2: And we want to make sure that parents feel like 159 00:07:24,320 --> 00:07:28,000 Speaker 2: they have a sense that they can have peace of 160 00:07:28,120 --> 00:07:30,720 Speaker 2: mind with our app and services. One of the things 161 00:07:30,760 --> 00:07:33,960 Speaker 2: with teen accounts was really to put in place safeguards 162 00:07:33,960 --> 00:07:36,680 Speaker 2: that match up to parents' top concerns. We want them 163 00:07:36,720 --> 00:07:39,240 Speaker 2: to know that these safeguards are in place. We're not 164 00:07:39,400 --> 00:07:42,320 Speaker 2: moving these safeguards. The only person, if your teen's under sixteen, 165 00:07:42,320 --> 00:07:45,320 Speaker 2: who can move those safeguards is you. And then we've 166 00:07:45,360 --> 00:07:47,880 Speaker 2: also built in additional controls so that if a parent 167 00:07:48,000 --> 00:07:50,000 Speaker 2: knows of a specific issue that their team may be 168 00:07:50,120 --> 00:07:53,520 Speaker 2: dealing with, they have the ability to see who their 169 00:07:53,600 --> 00:07:56,920 Speaker 2: team is messaging with. They can see who's following them, 170 00:07:56,960 --> 00:08:00,080 Speaker 2: who's following their team. They can set time parameters. If 171 00:08:00,080 --> 00:08:03,000 Speaker 2: their team really struggles with sort of self regulation around time, 172 00:08:03,480 --> 00:08:06,160 Speaker 2: they can actually put in place a fifteen minute limit 173 00:08:06,360 --> 00:08:07,120 Speaker 2: for our circus. 174 00:08:08,040 --> 00:08:10,360 Speaker 1: Will you be able to do you think take your 175 00:08:10,440 --> 00:08:12,880 Speaker 1: way out of all of this to a point where 176 00:08:13,160 --> 00:08:16,000 Speaker 1: this is no longer a conversation and if there is 177 00:08:16,040 --> 00:08:19,440 Speaker 1: a problem or an issue or fury and anger, the 178 00:08:19,600 --> 00:08:20,640 Speaker 1: take will solve it. 179 00:08:21,320 --> 00:08:25,680 Speaker 2: Yeah, that's a good question. I would say the tech 180 00:08:25,760 --> 00:08:29,240 Speaker 2: can solve a lot of pieces, but I don't think 181 00:08:29,400 --> 00:08:32,240 Speaker 2: it can solve everything, which is one of the reasons 182 00:08:32,280 --> 00:08:35,240 Speaker 2: that we're pushing for this legislation, because we think it's 183 00:08:35,320 --> 00:08:38,920 Speaker 2: the combination of legislation that immediately gives parents the right 184 00:08:38,960 --> 00:08:43,240 Speaker 2: to say yes or no with obligations, which can be 185 00:08:43,559 --> 00:08:47,360 Speaker 2: legislation or not, doesn't matter, but obligations that parent that 186 00:08:47,840 --> 00:08:50,200 Speaker 2: apps put in place the right safeguards, which is why 187 00:08:50,280 --> 00:08:54,079 Speaker 2: we've built teen accounts with the ability for parents to 188 00:08:54,240 --> 00:08:57,400 Speaker 2: see and have controls. It's that combination that I think 189 00:08:57,559 --> 00:08:59,080 Speaker 2: will provide us with a path. 190 00:08:58,880 --> 00:09:02,839 Speaker 1: Out broadbyst I question bad actors generally, How big a 191 00:09:02,960 --> 00:09:07,480 Speaker 1: problem are bad actors using things like AI looking to 192 00:09:07,600 --> 00:09:09,359 Speaker 1: cause trouble on your platforms? 193 00:09:10,000 --> 00:09:13,080 Speaker 2: Bad actors are adversarial, so they're always trying to find 194 00:09:13,160 --> 00:09:16,040 Speaker 2: ways around the safeguards that we've put in Are you winning? 195 00:09:16,720 --> 00:09:20,520 Speaker 2: I actually think AI provides a really great opportunity for 196 00:09:20,760 --> 00:09:24,240 Speaker 2: us more than for the bad actors, in the sense 197 00:09:24,320 --> 00:09:28,319 Speaker 2: that we are able to train bad the AI to 198 00:09:28,600 --> 00:09:32,439 Speaker 2: sort of identify their behaviors in much more sophisticate, caated 199 00:09:32,520 --> 00:09:35,280 Speaker 2: ways that are much less predictable for them to sort through. 200 00:09:35,320 --> 00:09:38,320 Speaker 1: An gain because here's my personal question to you. I've 201 00:09:38,360 --> 00:09:40,439 Speaker 1: got a profile in this country, and I've been on 202 00:09:40,559 --> 00:09:44,080 Speaker 1: the end of any number of scams, and I look 203 00:09:44,160 --> 00:09:46,000 Speaker 1: at those and I think, if you can't see that, 204 00:09:46,120 --> 00:09:48,880 Speaker 1: you're an idiot. Because they're not that good, and yet 205 00:09:49,240 --> 00:09:51,280 Speaker 1: people fall for them all the time. I'm assuming that's 206 00:09:51,320 --> 00:09:52,120 Speaker 1: only going to get worse. 207 00:09:53,920 --> 00:09:56,559 Speaker 2: They will get the bad actors because their official will 208 00:09:56,600 --> 00:09:58,720 Speaker 2: get will always try to get better and get around 209 00:09:58,760 --> 00:10:01,839 Speaker 2: our systems. But the technology gets much more sophisticated. So 210 00:10:02,160 --> 00:10:05,560 Speaker 2: for example, and sort of that clickbait of popular public figures, 211 00:10:05,840 --> 00:10:09,679 Speaker 2: we're now using these facial recognition tools and testing them 212 00:10:10,080 --> 00:10:11,080 Speaker 2: to stop that. 213 00:10:11,160 --> 00:10:17,360 Speaker 1: Type of skin Do you find your job enriching or problematic. 214 00:10:16,920 --> 00:10:21,600 Speaker 2: In this dying age? I, you know, I think there 215 00:10:21,600 --> 00:10:26,360 Speaker 2: are very few places where I would have the kind 216 00:10:26,400 --> 00:10:29,959 Speaker 2: of opportunity that I have at META to put in 217 00:10:30,120 --> 00:10:34,480 Speaker 2: place tools and safeguards that are helpful to parents. Before 218 00:10:34,600 --> 00:10:38,720 Speaker 2: joining META, I actually was a middle school teacher, I taught, 219 00:10:38,880 --> 00:10:41,640 Speaker 2: I have a law degree I had, I was working 220 00:10:41,679 --> 00:10:44,280 Speaker 2: for a regulator of thinking about these things. Most of 221 00:10:44,360 --> 00:10:46,800 Speaker 2: my career has been spent a lot of time on 222 00:10:47,200 --> 00:10:49,480 Speaker 2: thinking about parents and families and how do we support them, 223 00:10:49,800 --> 00:10:52,079 Speaker 2: And you can do that at great scale at META, 224 00:10:52,200 --> 00:10:55,200 Speaker 2: and so I feel and they're very supportive of Meta 225 00:10:55,240 --> 00:10:55,680 Speaker 2: of that work. 226 00:10:55,760 --> 00:10:57,679 Speaker 1: Fantastic. Nice to mate you and you must you must 227 00:10:57,720 --> 00:10:59,800 Speaker 1: come in again, not to see you and take any 228 00:11:00,080 --> 00:11:01,360 Speaker 1: Davis Haud have met it this morning. 229 00:11:01,880 --> 00:11:04,719 Speaker 2: For more from the Mic Asking Breakfast, listen live to 230 00:11:04,880 --> 00:11:07,920 Speaker 2: News Talks at B from six am weekdays, or follow 231 00:11:08,000 --> 00:11:09,520 Speaker 2: the podcast on iHeartRadio.