10 Policy Brief No. 44 Toda Peace Institute a request for comment. Indeed, the only major platform that seems to prioritise public transparency for its moderation process is Reddit, which, in 2018, gave New Yorker writer Andrew Marantz access to its internal moderation meetings.) Still, many employees at large platforms really do want to do the right thing and are struggling to develop solid, research-backed strategies that will work at unprecedented scale. (Some platform employees contributed to this policy brief, but wish to remain anonymous.) Meanwhile, in China, digital media platforms cooperate openly with the state. We know little about the philosophy informing government control of those platforms, but right now, it seems largely focused on preventing citizens from organising, and flooding the internet with positive posts. A 2013 study by King, Pan, and Roberts found that on Chinese social media, “Contrary to previous understandings, posts with negative, even vitriolic, criticism of the [Chinese] state, its leaders, and its policies are not more likely to be censored. Instead, we show that the censorship program is aimed at curtailing collective action by silencing comments that represent, reinforce, or spur social mobilisation, regardless of content.” A 2017 study by the same researchers found that many, perhaps most, of the propaganda posts on Chinese social media are simply “cheerleading” for China. Instead of outright censorship, this strategy focuses on soaking up citizens’ attention with nice-feeling posts about the loveliness of China and how lucky and grateful Chinese citizens are. This sort of activity does not just affect China’s internal politics. For one thing, citizens of other countries can and do use Chinese digital platforms. Indeed, many Chinese digital platforms are now growing aggressively outside China. The data they have gathered, and the control taken by the Chinese government, may now apply to millions of other countries’ citizens. And given that these platforms do not seem concerned about state censorship in China, there is no reason to think they will hesitate to cooperate with authoritarian governments anywhere else. 10. Positive Social Networks?: “Adaptive vs. Maladaptive,” or “ConnectionPromoting vs. Non-Connection-Promoting” The recent years of fever-pitch moral panic brought ethics to the forefront of the global conversation about digital media. Unfortunately, moments of moral panic are not ideal for long-term thinking. But hopefully good things will emerge from the sustained public pressure on tech companies we are now seeing. Most past research concentrated on showing that platform problems exist, because the problems were both frequently denied by tech companies and hard to understand, even for experts. Today, tech companies are trying different solutions (which I outline in my next brief, “Current Strategies and Their Discontents”). So is civil society, like NGOs whose operations take place on digital platforms. But when solutions are backed by research, the research largely is not transparent to the public. So, what can research tell us about building a digital platform that does good things in the