Facebook’s own technology spotted less than a fifth of bullying posts

Facebook wipes off two million bullying posts in three months… but its own technology spotted less than a fifth of them

  • Facebook’s own technology spotted just 15 per cent of abusive bullying posts
  • The firm takes action on two million abusive Facebook posts in just three months 
  • Prince William yesterday accused social media giants of allowing ‘bile and hate’ to remain online

Facebook’s technology is detecting just a tiny proportion of the abusive posts by bullies and trolls, it has emerged.

The firm said it had taken action on two million abusive posts in three months alone.

But just 15 per cent were detected by Facebook itself, with the company relying on users to root out the rest.

Artificial intelligence and algorithms fail to understand nuances in posts which might mean a message is abusive, the social network admitted.

Just 15 per cent of abusive posts are detected by Facebook itself. The stark figure came after the Duke of Cambridge called for social media firms to do more to fight cyberbullying [File photo]

The revelations came a day after Prince William accused social media giants of allowing ‘bile and hate’ to remain online.

A Facebook ‘transparency report’ published on Thursday night contained data relating to bullying for the first time.

Online trolling could be ‘especially harmful to minors’, the site admitted. But it added: ‘Bullying and harassment are highly personal by nature. 

In many instances, we need a person to report this behaviour to us before we can identify or remove it.’

Of the 2.1million abusive posts which were spotted from July to September, a user had got in touch to raise concerns in 85 per cent of cases, the report found. 

Prince William took a swipe at entrepreneurs such as Facebook founder Mark Zuckerberg, appearing to accuse them of putting profits over values. He said: ‘For too many families, phones and social media shattered the sanctity and protection of our home’

The small number spotted by Facebook tools means millions of similar posts are likely to remain up.

Online hate can lead to tragedy, as the Daily Mail reported yesterday. Writing in the paper, Lucy Alexander told how her son Felix took his own life at 17 after years of bullying. She said social media kicked in when Felix was around 14 ‘and the bullying became unstoppable’.

The stark figures in Facebook’s report come after the Duke of Cambridge called for social media firms to do more to fight cyberbullying.

William said: ‘For too many families, phones and social media shattered the sanctity and protection of our home.

‘I am worried that our technology companies still have a great deal to learn about the responsibilities that come with their significant power.’

He took a swipe at entrepreneurs such as Facebook founder Mark Zuckerberg, appearing to accuse them of putting profits over values.


  • Madeleine McCann’s parents post messages of ‘hope’ after…


    Zuckerberg KNEW about Russian interference in US election on…

Share this article

The US billionaire swerved a question about the royal’s comments during a media call on Thursday night, reeling off site rules instead.

Following a brief pause, he said: ‘We are going out of our way to proactively enforce the policies using this combination of AI systems and just dramatically growing the review team.

‘There’s also a big issue, on social networks, and in media more broadly, which is that people are just more likely to engage with more sensationalist and provocative and sometimes more negative content. We think that is bad.’

Earlier in the conversation, he had likened protecting the site to tackling soaring crime.

He said: ‘No one expects crime to be eliminated completely, but you expect things will get better over time. When you connect more than two billion people, you’re going to see all the good and the bad of humanity. 

‘These really are not issues that you ever fix. They are issues that you manage over time – and we need to get them to a good level.

Online trolling could be ‘especially harmful to minors’, the site admitted. But it added: ‘Bullying and harassment are highly personal by nature. In many instances, we need a person to report this behaviour to us before we can identify or remove it’ [File photo]

‘But the expectation that we should have is that there are going to be ongoing content issues.

‘If we are giving people a voice, some people are going to be posting bad stuff.’

The transparency report also revealed the reams of disturbing material being posted on Facebook.

From July to September nine million posts containing sexualised or abusive images of children were taken down. Facebook’s technology was able to find 99 per cent of them.

The site is also still being used by scores of online trolls and ‘bad actors’ hiding behind fake accounts, with 1.5billion taken down in the last six months alone.

The figure equates to half of the site’s monthly user base.

Source: Read Full Article