• by Fern Siegel , September 16, 2020

Tweets are posted, create controversy, and are taken down. That’s been the case for many of Donald Trump’s more provocative tweets — those spreading pandemic misinformation or retweeting White supremacist propaganda.

But the practice can be applied by Twitter, Facebook and Instagram for any violation of company policy.

For example, Facebook removed 7 million posts pushing dangerous COVID-19 misinformation from its site and Instagram between April and June. (Example: Trump posted an interview he gave to Fox News falsely claiming children are “almost immune” to coronavirus.)

Facebook also took down 22.5 million posts for violating its hate speech-rules in that time period, reported The Washington Post in mid-August.

What happens to the excised posts?

Human Rights Watch recommends social-media networks archive hate speech and other illegal posts after taking them down so they can be used as evidence in prosecutions.

Platforms are becoming more active in removing undesirable content, the New York-based group said in a Sept. 10 report. (HRW is a global nongovernmental organization that advocates on behalf of human rights. It is a founding member of the International Campaign to Ban Landmines, which received the 1997 Nobel Peace Prize.)

“This content also helps media and civil society document atrocities and other abuses, such as chemical weapons attacks in Syria, a security force crackdown in Sudan, and police abuse in the United States,” HRW wrote.

"Social media content, particularly photographs and videos, posted by perpetrators, victims, and witnesses to abuses, as well as others has become increasingly central to some prosecutions of war crimes and other international crimes," the organization added.

HRW recommends developing mechanisms to archive material removed because it promotes or incites violence, such as the gunman who livestreamed his attack on two mosques in Christchurch, New Zealand, in March 2019.

The worry is artificial-intelligence systems are taking down content before police can examine it or know it exists.

To address that issue, the Human Rights Center at the University of California, Berkeley, has an upcoming report —“Digital, Lockers: Options for Archiving Social Media Evidence of Atrocity Crimes” — that studies the possible archiving models for preserving material, creating a typology of five archive models, assessing strengths and weaknesses.

The concern is valid — since job one of any civilized society is the protection of its citizenry. In a social-media driven world, preserving evidence of serious crimes or dangerous misinformation is one way to help law enforcement, the courts and journalists better police, protect and inform people — and hold the guilty accountable.