Evidence timeline

c. Dec 2017Internal documentReportYouth Empathy Framing

"[O]ur product exploits weaknesses in the human psychology to promote product engagement and time spent"

Addiction & problematic useBody image & eating disordersBullying & harassmentLikes+3

"Young teens (~13) don't feel safe using their real name, identity or any clues that could lead to strangers finding them due to privacy and safety concerns." "PAC is keenly aware that prevalence rates for bullying/harassment is higher in teens than adults and that teens might require different forms of support." "9/10 teen girls say they are unhappy with their bodies; One study shows that girls express a heightened desire to change their appearance after spending time on FB." "Product features that are designed to exploit insecurity, or provide a dopamine rush (likes, notification, the pull-down-to-see, the infinite scroll, etc), to increase time spent, are inherently at odds with well-being and take away from people's ability to consciously focusing on activities that add value to their lives." "Because our product exploits weaknesses in the human psychology to promote product engagement and time spent [and that because of our reliance on AI, we cannot currently control for when content is being strategically leveraged to manipulate the opinions and moods of individual people] if we are committed to the wellbeing of individuals, we need to a) to alert people to the effect that the product has on their brain (for example: your view of yourself is affected by other people's highlight reel, etc) and b) provide them with tools to have more control over their own experience." [brackets in original]

4 Dec 2017Internal documentPresentationEarly Teen Illumination Qualitative Research

"Instagram Sells Itself: In 6th grade, age 11-12, teens get first phone and are 'invited' in one of two ways"

[JOINING INSTAGRAM]: [INSTAGRAM SELLS ITSELF]: "In 6th grade, age 11-12, teens get first phone and are 'invited' in one of two ways…" [BFF]: "My friend just got it and told me to download it." – ████, 12, LA [TOP ACCOUNTS]: "I want to see what my favorite YouTubers and soccer players are doing in real-life (off the screen… off the field)!" – ████ friend group, 14, S. Brazil

4 Dec 2017Internal documentPresentationEarly Teen Illumination Qualitative Research

12-Year-Old Instagram user: "Couldn't survive – need to see what everyone's up to - feel I am not missing out. I see what other 7th graders are doing that I should try."

Addiction & problematic use

[INSTAGRAM DRIVERS]: [EMOTIONAL BENEFITS / LIFELINE]: "IN CONTROL OF LIFE" • "I'm not alone" • "FOMO – not missing out" • "On trend" [EMOTIONAL BENEFITS / DISCOVERY AS ENTERTAINMENT]: • "Developmental need – scan word to find self" • "Curiosity peaked at age 13" • "Rewarding treasure hunt. Each layer leads to another" • "Escape after homework and before bed" • "I am more interesting 'I have more to talk to friends about'" [DRIVING BEHAVIORS / LIFELINE]: "My lifeline – Constant feed checking" "Couldn't survive – need to see what everyone's up to - feel I am not missing out. I see what other 7th graders are doing that I should try." - ████, 12, Chicago [DRIVING BEHAVIORS / DISCOVERY AS ENTERTAINMENT]: "Endless Treasure Hunt" "I go to bed at 9:30. I look at my clock and it's 1am!" - ████ 13, Chicago

21 Dec 2017 – 8 Jan 2018Internal documentMessagesRe: (Decision needed) Revert jewel filtering on US iOS decision

"we have abused the notifications channel as a company"

Notifications

Guillermo Spiller: "We're asking for your help to decide on a short-term trade-off between US DAP and notifications experience. We need to decide whether to revert the notifications filtering change inUS iOS." "Ultimately, the short-term trade-off centers around:" "A. Preserving a better notifications experience for people. While it looks like the downstream effects appear to be neutral (e.g., no change to comments), Core App thinks that a potential rollback is negative from a user experience standpoint. Filtering has eliminated 240 million jewel notifications/day (24%) for US iOS people. We estimate removing filtering in US iOS will increase a P90 person's daily notifications from 45 to 55/day, incremental push notifications will increase 306 million/day (17%) and incremental email will increase 130 million/day (38%) while we'll get an incremental 2.4 million negative feedback/day (24% more people hiding notifications)." "vs" "B. Recovering US DAP impact. We are expected to end the half 2.7M behind on US DAP, -41% of H2 goal; and trending to negative Q/Q US DAP growth. Reverting will increase iOS US DAP by 130k (2% of the expected growth for US in the half) and if we revert today, we would get +40k US DAP on reported Q4 numbers (given 30 days average), getting us from expected -200k to -160k Q/Q." "Do you think we should revert this change (option 1), or choose not to revert (option 2)?" Alex Schultz: "Fundamentally I believe that we have abused the notifications channel as a company. The biggest violators for me are things like app invites and group invites and historically we weren't able to remove them when my team pushed on removing them because they hurt metrics so I understand the feeling on not stopping for a metric BUT I believe metrics are a warning signal, not all of our users look like us and usage metrics (time, visitation, active users) are qualitatively v. different than using groups or using apps on our platform." "Are we really sure we've thought about the marginal user well enough in this work? Could these metrics declines be a flag there is something we aren't understanding?" Andrew Bocking: "just got clear input from Naomi that US DAP is a bigger concern for Mark right now than user experience. We have been pulled to see him this afternoon now at 3pm about this." Andrew Bocking: "I understand that NF/Notifications serve many interests, but we got a very clear and strong message from Mark that DAP (and specifically US DAP) is extremely important and we must change the trajectory from the negative one it's on now. It's an externally reported metric and therefore needs the right level of priority. I think it would trump video watch time and many other metrics in priority given this?"

2018Internal documentReportOverview: 2018 Instagram Ambassador Program

"Recruit and onboard teen tastemakers to act as our plug at local high schools within 5 key markets"

[Program Objectives]: "The ambassador program will aim to:" "Recruit and onboard teen tastemakers to act as our plug at local high schools within 5 key markets" "Drive product adoption through word of mouth for upcoming launch initiatives" "Create physical + digital challenges for ambassadors that engage peers and build buzz" "Whitelist ambassadors for approaching launches" "Ambassadors will act as our early adopters and help increase visibility" "Build rapport and strengthen relationships with community members" [Finding Ambassadors in Key Markets]: • "Target several cities from US City Target List for H1 2018" ◦ "New York" ◦ "Los Angeles" ◦ "San Francisco Bay Area" ◦ "Columbus" • "The ideal ambassador profile:" ◦ "13-17 years old" ◦ "Diverse" ◦ "Well-connected to peers, heavily involved and well-known within their school" [Using QP to target influential TEEN community members]: "QP will direct to a form, where ambassadors need to confirm that they are over the age of 13." [Sending NDA's + Getting parent permission]: "After teens respond to the ambassador invite email, we need to send a consent form and NDA" [Compensation]: • "We can pay US ambassadors up to $599 a year (due to tax reasons)" • "The Ambassador Program will pay teens $45 a month via Amazon Gift Card if they complete 3/5 challenges each month"

2018Internal documentReportH2/H1 2018 Youth Team Review

Facebook Youth Team outlines strategy to capture teens and tweens

Friend/Account Suggestions

"For Tweens, we need to develop a strategy to get them onto Messenger and Facebook." "We formalized our Youth Defense Task Force, tools and dashboards. This team focuses on building pipelines and systems to detect and respond to experiments and rollouts across Facebook that negatively affect teen metrics (teen MAP, DAP, TS and PS)." "Part of our strategy will be to aggressively experiment and discover use cases which currently work well for teens and to optimize and amplify them. There are likely many use cases which we haven't discovered yet." "One specific area we will focus a significant amount of effort on is school communities. We will work on the infrastructure to support school-related work (school graph completion)… Over an 18 month period we expect to drive an estimated +5% teen DAP (+2.5% teen MAP) from our high school work."

2018Internal documentReportH2/H1 2018 Youth Team Review

"Facebook is likely U13's first social media app (driven by parents)"

[Tweens]: • "The majority of US kids with phones use social media by age 9." • "Parents know about and connect to U13 on social media." • "Facebook is likely U13's first social media app (driven by parents)." • "U13s use multiple social media apps, but struggle to articulate FB's value."

2018Internal documentReportH2 2018 Communications Plan - We Get It: Meaningful Connections

"[We have] [f]ew allies on well-being issues as well as product or policy proof points"

Addiction & problematic use

[RISKS]: • "We're playing catch up and people might think we aren't doing enough. Per the New York Times 'If Facebook's underlying business model is based on how much time we all spend eating there, can the company ever truly resist the pressure to keep plying us with more cookies?' (January 2018)" • "Few allies on well-being issues as well as product or policy proof points" • "People don't know what controls are available or how to use them: research shows that people want controls but when we offer them they are not really used" • "Increasing negative focus from UK policy makers and media on the issue of addiction. Potential to spread."

2018Internal documentReportH2/H1 2018 Youth Team Review

"[W]e can estimate that there were 4M people under 13 in 2015 on IG. This represents around 30% of all 10-12 years old in the US."

[Instagram]: ◦ "In the US, using a new age affinity model specially developed for IG, we currently estimate there are 22.7M teens who are MAP. This is up from 11.4M estimated by the original lifestage model. The main driver for this step-change was the creation of a model for IG users that don't connect their account to FB. We learned that only 40% of teens have their IG account connected to FB, while adults have 77% of account connected." ◦ "Looking at people we predict to be 13 and 14 today, we can estimate that there were 4M people under 13 in 2015 on IG. This represents around 30% of all 10-12 years old in the US. Finally, assuming similar numbers today, 75% of US teens are MAP on IG."

Jan 2018Internal documentPresentationInstagram Well-Being Strategy Doc

"Model for how social media usage may be linked to suicide/self-harm"

Suicide & self-injuryBullying & harassment

"MODEL FOR HOW SOCIAL MEDIA USAGE MAY BE LINKED TO SUICIDE/SELF-HARM"

Jan 2018Internal documentPresentationInstagram Well-Being Strategy Doc

"8th graders who use social media for 10+ hrs/week are 56% more likely to be unhappy"

Addiction & problematic use

"A person's momentary happiness increases with low/ moderate Facebook usage" "However, these gains are reversed and actually decline below baseline with heavy usage (>3x/day)" "Risk of unhappiness due to social media use is highest for youngest teens" "8th graders who use social media for 10+ hrs/week are 56% more likely to be unhappy" "Teens who used social media every day are 19% more likely to get less than 7 hours of sleep" "Teens who visit social networking sites every day more likely to feel lonely" "Younger teens (8th graders) who are the most active on social media are also those who are at the greatest risk of developing depression"

2 Jan 2018Internal documentMessagesRe: Rob's Youth Update - Sep 7

Robert Chen: "our rivalry with YouTube (in combination with our former focus on the time spent metric) has led Facebook to take self-destructive actions such as doubling the amount of video watch time in 2017"

• "'The future is already here — it's just not very evenly distributed.' Is Facebook missing the future by not optimizing for teens in high GDP countries?" • "Including or not including YouTube in the kmeans clustering didn't change the results. This suggests that YouTube is different from the other apps (i.e., Facebook, WhatsApp, Snapchat, Instagram, Messenger) in the clustering" ◦ "I have come to believe more strongly that our rivalry with YouTube (in combination with our former focus on the time spent metric) has led Facebook to take self-destructive actions such as doubling the amount of video watch time in 2017"

25 Jan 2018Internal documentMessagesIG Teens Team FYI

"We can do a reverse infer to find out people who goto xx[High School], and potentially use it in some promotion flows"

Bullying & harassmentFriend/Account Suggestions

████: "I met with ████ from the TBH team earlier this week to pick his brain on teens... Here are the highlights: 1. Their growth strategy was to go school by school. They would create an IG account with a bio saying 'TBH coming to your school' and follow people who go to the school. They would get the list of people by crawling profiles and followers. Once they got enough followers, they would message, post, etc to get people to install the app. 1. High schoolers love anything that is exclusive to their high school community. He thinks community is more important than the actual product. 1. He thinks the directory feature is enough to get teens excited. A primary job TBH fulfilled was giving teens a way to see everyone at their school. 1. Bullying, sketchy people, and hacking will happen very quickly and will be a constant battle that we'll have to invest in. Bullying will even happen if we just have a directory. People will create accounts with usernames that make fun of people and join the network. Some integrity measures they used: 1. Requiring phone number verification. They ended up disabling VOIP phone numbers because of abuse. 1. Limiting the number of times you can switch high school. This helped, but many people would write in and ask if they can switch back. 1. Removing people who said they belonged to a HS but were not in the contacts of other people at the HS. 1. They found chat to be a good way for people to report abusive and bad activity since it's real time and you can assure the reporter that action will be taken." ████: "Re 1) Really interesting growth strategy using follower and their friend graph. Similarly, our FB ML school inference model also take friend graph into consideration. We can do a reverse infer to find out people who goto xxHS, and potentially use it in some promotion flows" ████: "i think this will work really well!"

Feb 2018Internal documentReportFeb 2018 Youth Small Group Update

"We're investing in aggressively driving high school profile completeness"

[High School Communities]: "[W]e learned that increasing high school profile completion drives engagement and DAP, as it makes it easier for Teens to message each other and join peer-heavy groups. We're investing in aggressively driving high school profile completeness this half."

8 Feb 2018Internal documentReport2018-02-08 - Schoooools Review

"[O]ne of the key goals of the Youth Growth team is to get teens onto Facebook and to stay there."

[Overview]: "The teen population on Facebook is decreasing and so one of the key goals of the Youth Growth team is to get teens onto Facebook and to stay there. To do so, the Youth Growth team is building products centered around High School Communities."

27 Feb 2018Internal documentPresentationIG Camera/Platform

"Build a daily behavior by giving Teens reasons to check the camera everyday through scalable new content"

Addiction & problematic useFilters

"STRATEGIC GOALS • Integrate platform into IG to see if AR effects can get strong product market fit with FCI Teens by tapping into IG teens community and cultural moments (this is something we haven't been able to achieve with our internal effects) • Target a list of ~50 teen creators, influencers or partners that resonate with FCI Teens through a closed beta in H1 and focus on key teen use cases: being funny, looking good, being part of a pop culture trend or moment • Measure success of beta program through participation of third party camera effects by FCI Teens; once product market fit is achieved with this demographic, quickly open the platform to scale quickly to reach daily use case • Maintain IG levels of quality and craft with third party effects through partner and creative guidelines" "PARTNERSHIPS BETA PROPOSAL • Prioritize authentic content that will resonate with FCI Teens • Implement a followership model where Partners get free distribution to their followers but need to pay for inorganic reach • Focus on key use cases: be funny, look cool / pretty, or be part of a pop culture trend or moment • Subsidize any or all Beta effects in order to set the right tone at launch • Partner with IGBP to understand business objectives of brands for inclusion in H2 • Leverage key moments: F8 in May and VidCon in June" "INSTAGRAM, TEENS, AND PARTNERS BENEFIT FROM THE PLATFORM Instagram: • Increase Camera Engagement in order to drive more sharing • Build a daily behavior by giving Teens reasons to check the camera everyday through scalable new content • Tap into communities and fan-bases that are thriving on IG • Help Teens make their friends laugh through funny tools • Help users feel comfortable sharing their face at any time Teens: • Use tools to be funny, make friends laugh, be entertained, or look cute/pretty • Crave unique creative tools they cannot get anywhere else • Want richer, more dynamic AR experiences that change frequently Partners: • Gain new channels for distribution and promotion through targeted impressions • Reach a large, highly engaged audience • Promote new products or self"

13 Apr 2018Internal documentMessagesRe: Rob's Youth Update - Sep 7

"We are aligned on the importance of young teens/tweens for winning teens longer term"

[What I will do]: • "Locations data: By 4/19, share normalized teen/adult FB location feature usage data" • "Branding: Connect with ████ on branding work which targets young teens/tweens (4/16)" • "Snapchat Teen Action Panel: Share update from MINT/ ████ on accelerating recruitment of US iOS teens for the action panel (by 4/19)" [Takeaways from the Competitive Landscape Review on 4/12]: • "We currently consider Instagram a competitor in many respects (e.g., teens social sharing) but also collaborate (e.g., high school data) with them" • "We are aligned on the importance of young teens/tweens for winning teens longer term" • "For Big Bets focused on the US, iOS is the most important platform to build for, given the high iOS market share" • "We want to explore authority figures and influencers as ways to onboard more teens more rapidly. These people are as varied as club leaders, student body leaders, teachers, and celebrities (e.g., Kim Kardashian)" • "We agree on the importance of brand and branding, especially with regards to young teens/tweens" • "High School, Fox Mode, and Close Friends are our current attempts at creating a separate teen space" • "We want to track whether IG has any softness in the younger age cohorts"

19 Apr 2018Internal documentMessages"Well-being: Youth and Technology" (Task T27934737)

████: "the issue with U13s is a study that we're sponsoring publishing findings related to U13 using FB products other than MKids. even without PII tied to the study, i think there's risk of the FTC picking up on it and asking us (or UW) for data about the findings"

"hey ████ - first off, my apologies for not being at the meeting on monday due to PTO. i realize my absence is now causing thrashing that i could've avoided. i've read the documents. everything looks good from a 13+ POV (thanks ████!). the issue with U13s is a study that we're sponsoring publishing findings related to U13 using FB products other than MKids. even without PII tied to the study, i think there's risk of the FTC picking up on it and asking us (or UW) for data about the findings"

20 Apr 2018Internal documentMessagesRe: Rob's Youth Update - Sep 7

Robert Chen: "We have an acquisition problem for young teens"

Notifications

• "Locations Data: US teens overindex by 2.3X on Messenger live location sends compared to adults" ◦ "Implication: This suggests that there may be an opportunity to build more location products for teens" "Implication: We have an acquisition problem for young teens. Strongly negative brand likely plays into this." • "Why do people open Facebook? - from ████" ◦ "Younger people come for notifications around messaging" ▪ "Implication: Let's figure out how to put messaging in the main app for teens"

23 Apr 2018Internal documentReportHigh School Directory - Product Messaging & Launch Doc. [working draft]

Instagram's Teen Ambassador Program used teen influencers to market to peers at school

"Support pilot program and medium-volume launch of Nametag high school networks, the first product just for teens, and leverage it as a first-ever opportunity anopportunity to experiment with teen-first, never-before-tested hyper-targeted house ads and partner/community integrations outreach tactics optimized for US high schoolers." "Give teen ambassadors early access to product and work with them to build excitement at local high schools" "New Teen Ambassador Program gives us newfound access to teens – let's leverage this by having 'the plug' get access to the product before launch and have them grassroots tell their friends about it (in class, at assembly, after school, etc.)"

May 2018Internal documentPresentationAge Modeling: a ripe opportunity

"We have chosen not to perfect our age modeling to avoid the legal risks that come with actual age knowledge"

"We use two forms of age modeling, Life Stage and Age Affinity. For legal and policy reasons, we only acknowledge using the less accurate one (Life Stage) and have blocked perfecting the better one (Age Affinity). The scales have tipped and the benefits of perfecting the model and speaking about it openly outweigh the risks. The time is ripe to perfect the model and be transparent." "We quietly use Age Affinity across a number of different vectors – creating transparency risk." • "Eg, News Feed Ranking, PYMK, Analytics" "We have chosen not to perfect our age modeling to avoid the legal risks that come with actual age knowledge – hurting product efforts and diminishing value of safety efforts." "We have leaned into the imperfection of our model to push back against regulator demands – harming policymaker trust." "Legal Costs Increases platform liability - once we know, we're responsible"