Meta said on Friday it would overhaul its moderation system after a wave of criticism that it treated VIP posts as privileged over regular users.
The parent company of Facebook, Instagram and others said it would implement most of the recommendations of an independent oversight panel on its policy to exempt mass users from some content moderation laws.
However, Meta rejected some proposals to increase transparency and reduce bias in the offices of public figures.
In November 2021, the Meta verification system came under fire after a Wall Street Journal report claimed that the system was blinding some high-level users.
Following the report, the watchdog, which is funded by META and whose members say it operates independently, called for an overhaul of the controversial verification system, which gives special treatment to messages from millions of celebrities, politicians and Instagram accounts. Other influencers who let them post content they love. This is against company policy.
According to the report, the number of high-profile accounts registered with the program rose to 5.8 million users in 2020. After the report was released, Meta's oversight board said Facebook had failed to provide it with key system details.
At the time, Meta said it was "fair to criticize the system," but the review system was created to improve accuracy when editing content that "may require deeper understanding."
Meta said today that it will improve its criteria for distinguishing between accounts included in the Expanded Company Screening Program and human rights accounts, and explain these distinctions in detail to the Oversight Board and Transparency Center.
Meta will also improve its process to temporarily remove or hide potentially harmful content until the end of the additional review period, and the company will ensure that content reviewers have the appropriate language and regional skills "as far as possible."
Meta rejected a recommendation from the Oversight Committee that social media companies differentiate between Pages and accounts of public figures, who receive special treatment in their content moderation policies. The company said such public identifiers can make accounts "potential targets for bad actors."