Mr. Mark Malone:

I am happy to address that. A few things could be done on the algorithms, but perhaps it would be useful to give some context about what we see. The Deputy spoke about having one-to-one conversations, which is powerful and effective. We measured some YouTube views on explicitly anti-migrant and anti-refugee content that was created by a tiny number of accounts. It was seven accounts. Those accounts amassed more than 1 million views from the start of December to mid-January. When we put that number of views being shared in context - we are not entirely sure the views were all in Ireland - it allows the algorithms to promote that content. A gaming of content is going on that is facilitating it to be at the top of the feeds of people who are not even searching for this reactionary content. A concern, particularly with YouTube, is that some of these channels are monetised. We have seen live feeds from protests, including calls for the burning down of accommodation centres, for which the individuals who were broadcasting it were making money on the YouTube channel. YouTube takes a 30% cut of any money raised. There is a structural problem across the platforms. It is in Meta as well. FRO has trusted flagger status on all of the main platforms. Each of them has regularly failed to remove content we have reported. The problem is not that the platforms do not have policies but their ability, or whether they are choosing, to uphold them. When we look at what we all see on social media and witness - our colleagues across the board have spoken about the spread of disinformation - it is not plausible that these companies do not know it is happening.