Facebook using artificial intelligence to prioritise reported content

Facebook on Tuesday said it has stepped up the use of artificial intelligence (AI) to "prioritise reported content", a move that will help the social media giant take action faster on harmful and violative content.

Facebook, which has 1.82 billion daily users globally, has drawn flak in the past for its handling of hate speech on the platform in India, which is among its biggest markets.

Facebook Product Manager (Community Integrity) Ryan Barnes said the company is using AI to prioritise reported content, and that this prioritisation is important to help its over 15,000 reviewers.

She explained that the prioritisation is important for four reasons -- not all harmful content is equal, some enforcement decisions are complex, people do not always report harmful content and the reports aren't always accurate.

Speaking to reporters in a virtual briefing, she said the company has moved from relying on user reports alone to add use of technology to help aid the process.

"We've moved away from just reviewing things chronologically to using AI to help us prioritise what we review. We have looked at severity which has been a factor in our prioritisation, but now we have other factors such as virality, severity and likelihood of violations," she said, adding that this will help "act" on reports faster.

Barnes said the community integrity team is focussed on reducing the prevalence of bad experiences by taking action on violating content and abusive actors proactively and with fewer mistakes.

She added that over 95 per cent of such content is spotted by the company's technology before anyone reports it to Facebook.

Facebook Engineer (Community Integrity) Chris Palow said using AI will help in getting to the "most harmful content faster" and give human review teams more time to spend on complex decisions.

He added that this will also help identify new trends, and respond to people attempting to post violating content.

Asked if AI can help in handling hate speech, Barnes said Facebook has no tolerance for hate speech on its platform.

"We realise this is an important issue, and we in fact think that that AI technologies can help with this in terms of how do we determine if a piece of content needs human moderation...depending on how viral that piece of content is or how severe that piece of content is or its likelihood of violating," she said.

Barnes added that the company is using and relying on its technology to bump up such content in terms of ranking as to what reports are reviewed first.

The company's decisions on 'dangerous persons' is a comprehensive process, including consultation with partners on ground and review of online and offline behaviour.

"We don't use AI to make decisions about designating people as dangerous individuals under our policy. But AI can help us find content that may help to inform that decision, but it's not actually making the decision outright," a company spokesperson said.

The spokesperson added that while the community guidelines are uniform globally, local context and cultural nuances are kept in mind as these policies are developed and enforced.