Meta & Children's Online SafetyAgency and Control Audit Comms Considerations
Raw text
Case 4:22-md-03047-YGR Document 2757-31 Filed 02/20/26 Page 1 of 9 AMENDED Exhibit 253 PLAINTIFFS' OMNIBUS OPPOSITION TO DEFENDANTS' MOTIONS FOR SUMMARY JUDGMENT Case No.: 4:22-md-03047-YGR MDL No. 3047 In Re: Social Media Adolescent Addiction/Personal Injury Products Liability Litigation Case 4:22-md-03047-YGR Document 2757-31 Filed 02/20/26 Page 2 of 9 Agency and Control Audit comms considerations What the audit is about: Dark patterns and "addictive"-like features rob users of their agency We want to systematically identify and resolve features that fall in this category. Identify where undesirable patterns are present throughout the Facebook app and website Create guidelines to integrate into Facebook Design System and associated design tools across the company Confront and revise the design where these patterns are identified See more: Project Plato PLAN: DESIGN AUDIT Perform a design audit of the platform with key external behavioral experts and with internal designers and researchers to identify features that may lead to undesired, habitual, or unwanted/low-value behaviors How we'll work with external expert: Share and iterate on our definition of characteristics of features that promote undesired or problematic behavior. • Engage with a third party/experts to conduct an independent audit and/or review of the features we have identified $ Form an alliance where the third party can vouch for the thoroughness and relevance of our approach for targeting the "addiction" daims Benefits • Proactively address the negative press • Claims that we are "sprinkling behavioral cocaine all over our products," accusations of using dark patterns to manipulate users' privacy, and other extreme views are pervasive in the media. We have an opportunity to present our response to these claims and preempt any regulations • We can highlight that we are paying attention to addressing this important problem and applying rigorous research methods to understand the validity of the claims • Transparency • Credibility • By partnering with external design experts and/or academics, we can show that we are taking a comprehensive, unbiased approach to identifying potential sources of problematic experiences for our users. * External allies • We can use this process to form allies in the broader tech and academic communities, particularly among some of our most vocal critics (e.g., Tristan Harris). We can demonstrate that we are taking user concems and findings of problematic use seriously. Someone like Tristan already has a lot of thoughts and connections in this space, so we can leverage his expertise. Ideally working with an external expert on this could make them see we're taking this seriously and they may become vocal advocates of ours we can point people to. • Possibility to improve user well-being and FB sentiment • By sending the message to the public that we support users' self-control, build products that are in the users' best interests, and respect users' autonomy and agency, we could see a boost in both user well- being associated with platform use as well as FB sentiment metrics. With well-being-related tracking metrics forthcoming, this could be something to follow and present as a benefit to any engagement drop-offs that are seen. We also could consider partnering with the Consumer Marketing team as well as Goodwill team on this. Risks (and ideas for how to mitigate them) What if we disagree with the validity of the recommended changes? • We can frame this audit as identifying hypotheses that all need to be tested. HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-020-00256107 Case 4:22-md-03047-YGR Document 2757-31 Filed 02/20/26 Page 3 of 9 What if we agree with recommended changes, but for various reasons don't want to make them? • It is likely that there will be several possible solutions to address potentially problematic features that are identified as part of the audit. In our communications with external contributors as well as public narrative, we can focus on the need to further investigate the impacts of any features identified in the audit, and evaluate the extent and severity of negative impacts. It is worthwhile to consider how far we might consider going to make tradeoffs that prioritize user agency and that can simultaneously reduce engagement (unwanted engagement perhaps, from the users' perspective, but engagement nonetheless). • Changes could result in drop offs in engagement metrics (at least in certain markets). • We need to consider what our objective/data-driven signs of benefits would be from this effort (comms/public narrative, sentiment, user well-being) and get alignment of teams who may be negatively impacted at least in the short term. Possible benefits to doing this before we initiate the audit. • "Addiction" narrative: If we bring up the subject of problematic use/ addiction, our critics will talk about it, also. • This isn't new fuel, and the conversation is already going on (particularly related to addiction and privacy concerns). We can take ownership over our role in addressing this problem and frame the narrative. • Misunderstanding/cynicism: People could think we are doing the research in order to make Facebook "more addicting." • Consider doing proactive messaging that calls out the Hard Questions blog post, Facebook's principles/values, and upcoming well-being efforts. Don't call out the project as being about "dark patterns" or "addictive-like features" since that could sound like we are trying to learn how to make more features that manipulate users. Similarly, change product teams' goals and intermal narrative around making experiences "habit-forming" or "automatic." Relevant claims from the current external narrative How Technology Hijacks People's Minds Tristan Harris, Design Ethicist - Center for Human Technology Kernels of truth and possible action points, inspired by selected (more extreme) claims in the article: • Options presented on menus reflect nudges on behavior, and these should be in presenting valuable choices that are in keeping with users' best interests and intentions. Tristan says: "By shaping the menus we pick from, technology hijacks the way we perceive our choices and replaces them with new ones." • Intermittent, unpredictable rewards increase motivation for repeated behavior; reward timing should not be artificially made to be unpredictable, and options should be presented to organize or 'batch' rewards Tristan says: "Apps and websites sprinkle intermittent variable rewards all over their products because it's good for business... But now companies like Apple and Google have a responsibility to reduce these effects by converting intermittent variable rewards into less addictive, more predictable oneswith better design. For example, they could empower people to set predictable times during the day or week for when they want to check "slot machine" apps, and correspondingly adjust when new messages are delivered to align with those times." Platforms should not manipulate users into feeling like they are missing something that they cannot get at another point in time. Tristan says: "If I convince you that I'm a channel for important information, messages, friendships, or potential sexual opportunities - it will be hard for you to tum me off, unsubscribe, or remove your account - because (aha, I win) you might miss something important... Imagine if tech companies recognized that, and helped us proactively tune our relationships with friends and businesses in terms of what we define as "time well spent" for our lives, instead of in terms of what we might miss." ॐ Need for social approval is a powerful human drive; we should present options to allow people to opt out from moments of evaluation (e.g., pause or remove the option to receive tags/comments/reactions on certain posts - timeline review is a great example of this). Tristan says: "So when Marc tags me, he's actually responding to Facebook's suggestion, not making an independent choice. But through design choices like this, Facebook controls the multiplier for how often millions of people experience their social approval on the line." Limitless content should come with cues for pauses, completion, or active choice to continue. Tristan says: "Another way to hijack people is to keep them consuming things, even when they aren't hungry anymore. How? Easy. Take an experience that was bounded and finite, and turn it into a bottomless flow that keeps going.... News feeds are purposely designed to auto-refill with reasons to keep you scrolling, and purposely eliminate any reason for you to pause, reconsider or leave." HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-020-00256108 Case 4:22-md-03047-YGR Document 2757-31 Filed 02/20/26 * * Page 4 of 9 Notifications and messages could be set to come through according to the receiver's timeframe, instead of the sender's timeframe, and read receipts could be removed (or opt-in). Tristan says: "Interruption is good for business. It's also in their interest to heighten the feeling of urgency and social reciprocity. For example, Facebook automatically tells the sender when you "saw" their message, instead of letting you avoid disclosing whether you read it ("now that you know I've seen the message, I feel even more obligated to respond.")" Misdirection and roach motel are dark patterns that are detrimental for users' sense of agency and should be reduced/eliminated. Tristan says: Businesses naturally want to make the choices they want you to make easier, and the choices they don't want you to make harder.... apps can exploit people's inability to forecast the consequences of a dick. • Hypothesis: Grayscale makes apps less appealing, so this may make it easier to control your use; this could be used selectively. (Tristan Harris, cited in Change Your Phone to Grayscale to Combat Facebook Addiction) Don't Delete Facebook Just Be Smarter on Facebook David Murphy, Senior Tech Editor of Lifehacker Some takeaways: It can be helpful to break habits in order to control use: "Retrain your hand to stop typing in "www.fa..." every time you open a new tab." • Allow for deactivation without emotional manipulation: "I really, really hate Facebook's "these random friends of yours we pulled from your feed are totallilly going to be sad if you leave no really" begging." • Apps shouldn't provide easy options that might compromise privacy: "It's convenient to press a blue button and log into various sites and services with your Facebook credentials. But when you give other companies access to your Facebook data- which can be quite comprehensive- you're surrendering control of that information to entities that probably don't care very much about you or your privacy. Stop doing that." * Reconsider whether default options encourage selections that share personal data: "By default, other people who can see your info can bring it with them when they use apps, and your info becomes available to those apps. You can limit this as follows" Example of public response to previous Facebook acknowledgment of nuanced impacts of use, including potentially negative impacts Internal / External quotes about the Hard Questions blog post: "Is spending time on social media bad for us?": From Chris Cox (December 18, 2017): I wanted to tell you both how much I appreciate and respect the blog post you published this week. Reading the coverage it's even dearer how important it is that we be participating in this conversation, and you've put our voice out there in such a thoughtful and honest way. Thank you both for putting yourselves out there, and for doing the work to represent the company, and more importantly, the best interests of the people who use our products. I'm really proud of how this came off, and I hope you both are too. From (December 15, 2017): You guys - I know how hard this post was today- but I am reading through all the EOD coverage again and it is just awesome. This is EXACTLY how we need people to view us. Yes, we made some tough admissions, and there were some rough headlines. But that HAD to happen, and overall the coverage is that we did something extraordinary and that we took a massive risk and were HONEST. I am personally so grateful to you all for making it happen. This may be my favorite HQ post of all time. Thank you again for all that went into it. From at Morgan Stanley in his moming email (December 18, 2017; Chris, and other FB leads): shared this with HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-020-00256109 Case 4:22-md-03047-YGR Document 2757-31 Filed 02/20/26 Page 5 of 9 FB's corporate posture is evolving -- this is clear in both the company's willingness to discuss publicly if/how spending time on social media is 'bad' for us (see the company's blog here), but also how they have articulated their stance/ solution. You have to read the blog post to really digest the message here, but the incredibly short version is that 'passive' use of FB ("reading but not interacting with people") makes users feel worse while 'engaging' on the app ("sharing messages, posts and comments with close friends and reminiscing about past interactions") actually makes people feel better, per studies cited by FB. At a high level, their willingness to shine a light on critical research, particularly that which may not make FB 'look great, is part of the company's effort to let people know how much time and attention they pay to the social ramifications and responsibilities of what they have created. They want people to know that they have a team of scientists analyzing the social implications of FB (both in-house and working with other leading research facilities) -- and if you think FB is sending your kids down the wrong path, then at least you should appreciate (some) that even they agree there needs to be a 'better Facebook' and that it is pledging to build one (that is welcome and new). This is strategically important because it keeps the PR message positivethrough some of these negative studies and articles. At the same time, they clearly want people to know the importance of active engagement (which they are focused on growing) which is "good" for society (in their studies) and frankly good for monetization too. It's a delicate subject that's not going away, but one FB seems to be messaging thoughtfully on (and a few communications missteps very early on). That is a 'good' thing to see. (FB) Farhad Manjoo (NY Times, December 15): This is one of those stories where what's being said isn't as surprising as who's saying it. Facebook's using a corporate blog post to point to independent research that shows its product can sometimes lead to lower measures of physical and mental well-being should be regarded as a big deal.... This gets to why an otherwise in-the-weeds blog post from Facebook's research team is so interesting. Though it is quite abstruse, the post, by David Ginsberg and Moira Burke, two company researchers, takes readers through a tour of the nuances on whether Facebook can be bad for you.... I'm more optimistic about it, because it's in line with an evolving corporate posture from the company.... To be sure, Facebook is putting its own favorable spin on these studies. Yet its willingness to shine a light on critical research, and its pledge to take the findings into account when it designs its products, has to be welcomed as something new." Contact Support Conversation History Quipbot#21 left the thread Oct 12 at 9:50 am Quipbot#21 Oct 12 at 9:45 am =====IMPORTANT====== This Quip file has been copied to Google Workspace (https://docs.google.com/document...) as part of the Quip2Google migration efforts. Quip is moving to Read-Only. Please work in the new Google file moving forward. Learn more https://fb.workplace.com/groups... Questions? Visit the Quip2Google Migration Feedback Group (https://fb.workplace.com/groups...). Note New edits in this Quip file will NOT be automatically synced to Google Workspace. Quipbot#21 added Quipbot#21 to the thread Oct 12 at 9:50 am Quipbot#21 added you to a document Oct 12 at 9:45 am HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-020-00256110 Case 4:22-md-03047-YGR Document 2757-31 Filed 02/20/26 Page 6 of 9 HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-020-00256111 Case 4:22-md-03047-YGR Document 2757-31 Filed 02/20/26 Page 7 of 9 HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-020-00256112 Case 4:22-md-03047-YGR Document 2757-31 Filed 02/20/26 Page 8 of 9 HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-020-00256113 < Case 4:22 md 03047 VCD Document 2757 21 Filed 02/20/26 Page 9 of 9 HIGHLY CONFIDENTIAL (COMPETITOR) META3047MDL-020-00256114