More Transparency and Less Spin

Meta has some explaining to do.

On January 7th, Meta announced a raft of sweeping changes to its existing policies, including changes to fact-checking and its enforcement of policies on harmful content.

CCDH analyzed these changes and found:

• Content enforcement could be halted in 97% of key areas including hate speech, bullying and harassment, and violence or incitement of violence.
• Changes to hate speech policy would be implemented worldwide & immediately
• Other major policy changes currently impacting only users in the US, would eventually be expanded beyond the US.
• Meta has failed to adequately explain to users why or how these changes will be implemented across their platforms.

An intro from CCDH CEO Imran Ahmed

In 2018, Mark Zuckerberg’s Facebook publicly accepted blame for allowing its platform to be used to “foment division and incite offline violence” that led to a genocide in Burma.¹ A sophisticated online propaganda campaign on Facebook conducted by the Burmese government, urging a national campaign of civilian slaughter, rape, and the mass displacement of entire communities, had been allowed to unfurl without consequences. 

When such things happen, people solemnly promise each other that they will ‘never forget’. For Mr. Zuckerberg, however, it appears six years was long enough. In January, Mr. Zuckerberg and his policy chief, Joel Kaplan, announced a new era for Meta, tearing up six years of slow, incremental change made in response not just to the Burmese genocide but a litany of sins that Meta had once acknowledged as its problem and had committed to solving. However, Mr. Zuckerberg and Mr. Kaplan failed to address what these changes would mean in practice in their breathless zeal to announce their change of heart. 

To quantify the impact of Meta’s newly announced policy platform, this new report by CCDH systematically reviews Meta’s own data to piece together what we know, what we can safely predict, and we identify important unanswered questions that need urgent answers from Meta. Our analysis shows that the company’s plans could affect 97% of its enforcement, which would result in a tidal wave of nearly 277 million pieces of hate speech and other harmful content flooding unabated onto the platform each year, threatening public safety, the integrity of our elections and democracy, as well as public health and the mental health of our children. 

For the average user, this will mean that, when scrolling through Facebook, Instagram, or Threads, they are more likely to experience a chaotic blend of facts, opinion, and lies once Meta’s proactive systems of detection of hate speech and misinformation are removed and its global fact-checking program is replaced by a copy of X’s flawed Community Notes. Users are also far more likely to come across dehumanizing, hate-filled lies once prohibited by Meta, such as: “immigrants are grubby, filthy pieces of shit”; “Black people are more violent than Whites”; and “Jews are greedier than Christians”. These aren’t hypotheticals. These are specific examples cited in Meta’s new internal guidance.² 

Meta should have asked themselves all the questions raised in this report before announcing the changes. We urge legislators, regulators, civil society, journalists, citizens, and users to ask Meta these questions – whenever and however they can. 

Imran Ahmed 
CEO, Center for Countering Digital Hate