Elon Musk's Twitter Is Making Meta Look Smart

It was April 1, 2022, and I was sitting in a conference room at a midtown Manhattan law firm at a meeting of the Meta Review Board, an independent body that reviews its decisions on the merits. And for a few minutes, it was as if despair took over.

It was about Meta's controversial Cross Check program, which approved the posts of some influential users: celebrities, journalists, government officials and others. This program has been secret for years, and Meta even misled the board about its scope. When details of the program were published in The Wall Street Journal , it was revealed that millions of people received this special treatment, which means they are less likely to have their posts removed by algorithms or other users for violating the rules. To speak out against things like hate. . The idea was to avoid mistakes in situations where they would have more impact or spoil the Meta, depending on the speaker's preference. Internal documents revealed that Meta researchers had doubts about the feasibility of the project. Only after this revelation did Meta ask its board of directors to review the plan and advise what the company should do about it.

The meeting I attended was part of this demonstration. And the tone of the discussion made me wonder if the Meta board would suggest shutting down the program entirely in the name of fairness. "Politics should be for everyone." shouted one of the council members.

This did not happen. The social media world got a break from watching Elon Musk's Twitter train wreck this week when the Oversight Board finally released its cross-examination report from the Meta segment. (He never provided the list, at least not until someone took a closer look at the record to see who had received special permission to prevent murder.) The findings were stark. Meta said the goal of the program was to improve content decision-making, but the board of directors decided it was more about protecting the company's business interests. Meta has never established processes to monitor the program and determine whether it is fulfilling its mission. The lack of transparency to the outside world was appalling. After all, the Meta was too often unable to perform quick personal actions, so these messages were quickly dismissed. There were a lot of things like that that the meta team couldn't handle. They are often awake for days before getting secondary attention.

The first example that appeared in the WSJ's original report was a post by Brazilian soccer player Neymar in September 2019, which published a sexually explicit image without the subject's consent. Because of the special treatment she received for being part of the Cross Check elite, the photo, which blatantly violates the policy, garnered more than 56 million views before it was finally removed. A program designed to reduce the impact of content rating errors has increased the impact of scary content.

However, the board did not recommend disabling Meta cross-validation. Instead, a major overhaul was required. The reasons are in no way an endorsement of the program, but an acknowledgment of the devilish difficulty of content moderation. The context of the Review Commission's report was a desperate view that the situation could be remedied. Meta, like other platforms that give users a voice, has long emphasized growth over caution, offering large volumes of content that are expensive to police. Meta spends millions on moderation, but still makes millions of mistakes. A serious reduction in these errors costs more than the company is willing to spend. The idea of cross-validation is to reduce the error rate in the works of the most important and distinguished people. When a celebrity or public figure spoke to millions of people using their platform, Meta didn't want to hang his head.