Twitter taps into machine learning to stop abusive content

Read time 2min 20sec
Jack Dorsey, Twitter CEO.
Jack Dorsey, Twitter CEO.

Micro-blogging firm Twitter says technology solutions such as machine learning are helping the social media company to better detect malicious content on its platform.

Twitter, which released its bi-annual transparency report yesterday, says by using technology, 65% of the abusive content it actions on is surfaced proactively for human review, instead of relying on reports from people using the platform.

The social media company says from July to December 2020, through the use of machine learning, there was a 142% increase in accounts actioned, compared to the previous reporting period – 964 459 in total.

For SA, the report comes as government is concerned about the use of social media as a “dangerous tool” to mobilise ongoing riots.

Social media pundits highlighted the ways in which online platforms Twitter, Facebook, Instagram and TikTok have been used by rioters to co-ordinate looting and trigger the violent attacks.

The unrest has claimed the lives of 72 people, with over 1 700 suspects arrested in KwaZulu-Natal and Gauteng.

In its report, Twitter says it has begun enforcing its hateful conduct policy against content that incites fear, which has since been expanded to include content that dehumanises on the basis of race, ethnicity, or national origin.

Turning to COVID-19 misinformation, Twitter says: “As the COVID-19 pandemic evolves around the world, we continue to take enforcement action on misleading information about COVID-19, particularly that which puts people at risk and could lead to harm.”

From July to December 2020, Twitter challenged 10 320 924 accounts.

“The account challenge number represents a large volume of proactive anti-spam challenges we issued, targeting platform manipulation focused on COVID-19 discussions. We rolled out a set of proactive challenges specifically focused on COVID-19. We suspended 597 accounts, and removed 3 846 pieces of content,” it says.

“Since introducing our COVID-19 guidance last year, to the time of this publication, we have challenged 11.7 million accounts, suspended 1 496 accounts and removed over 43 010 pieces of content worldwide.”

On legal demands for content removal, during this reporting period, Twitter, which is led by CEO Jack Dorsey, says it received 38 524 legal demands to remove content specifying 131 933 accounts.

“We withheld or otherwise removed some or all of the reported content in response to 29% of these global legal demands – 11 091 total.”

The accounts of 199 verified journalists and news outlets from around the world were subject to 361 legal demands − a 26% increase compared to the previous reporting period.

According to Twitter, although there was a 9% decrease in the number of legal demands it received, compared to the previous reporting period, these requests sought removal of content from the largest number of accounts to date in a single reporting period.

See also