Filenews 10 October 2021 - by Theano Thiopoulou
Fake news, disinformation and hate speech are expressed online, resulting in various initiatives being taken at European Commission level to reduce the incidence.
The online platforms Facebook, Twitter, TikTok, Microsoft and Google cut off any news they consider false. The eight new potential signatories joined the process of revising the Code of Conduct on Disinformation. The list of new potential signatories includes video platforms such as Vimeo, new types of social networks such as Clubhouse, and providers of advertising technologies such as DoubleVerify, as well as organizations that provide expert expertise and technical solutions to combat disinformation, such as Avaaz, Globsec, Logically, NewsGuard and WhoTargetsMe.
What goes through the internet
The European Commission published last week the results of the sixth evaluation of the Code of Conduct on combating illegal hate speech online. The results show a motley picture, as IT companies reviewed 81% of notifications within 24 hours and removed an average of 62.5% of content. These averages are lower than those recorded in 2019 and 2020. While some companies have improved, the results for others have clearly deteriorated.
69% of content requesting murder or violence against specific groups was removed, while 55% of content using slanderous words or images targeting specific groups was also removed. In contrast, in 2020, the corresponding results were 83.5% and 57.8%. Sexual orientation is the most frequently cited hate speech (18.2%), followed by xenophobia (18%) and anti-gypsy positions (12.5%).
Disinformation News
The code is the first such framework in the world to set out the commitments of platforms and industry to fight disinformation, according to the European Commission. The revision of the code builds on the Commission's guidance document adopted in May, which sets out how the current code should be strengthened to provide a robust response to disinformation. As demonstrated by recent election campaigns, significant additional efforts are needed to reduce the flow of harmful disinformation.
The current code was a good first step in this direction, but the Commission's assessment in 2020 revealed significant shortcomings. These include inconsistent and incomplete implementation of the Code, across platforms and Member States, gaps in meeting the Code's commitments, the lack of an appropriate monitoring mechanism, including key performance indicators, and limited involvement of stakeholders, in particular from the advertising sector. The Commission expects the signatories to strictly adhere to the guidance document when revising the Code to ensure that it meets expectations for a strong EU instrument.
Social media manipulation
Platforms have stepped up their efforts to identify cases of social media manipulation and malignant influence functions or coordinated inauthentic behaviour. Although the platforms identified a large amount of content, including false information related to COVID-19, they did not detect coordinated disinformation actions with a particular focus on the COVID-19 pandemic. In addition, the reports highlight strong actions taken to limit the flow of ads on third-party websites that eliminate misinformation about COVID-19, while providing free COVID-related ad space to government and health organizations. The reports contain quantitative data illustrating the impact of these policies.
For example, Google search highlighted articles published by EU audit organizations that generated more than 155 million impressions in the first half of 2020. YouTube, since the beginning of the year, has been displaying dashboards of information linked to global and locally relevant health employees on its homepage and on boards that appear in videos and searches about COVID-19. In total, these panels have served over 300 billion views around the world.
Microsoft's LinkedIn sends interested members a European Daily Rundown, which is a summary of the day's news written and curated by experienced journalists. The European Daily Rundown is distributed to members in all 27 EU Member States and has an approach of around 9.7 million users in the EU.
Facebook and Instagram are directing over 2 billion people worldwide to resources from the World Health Organization and other health authorities, with more than 600 million people clicking to learn more. In April, in the EU more than 49 million people visited the COVID-19 Information Centre.
Over 160 million people have visited Twitter's covid-19 management pages, over 2 billion times. Such pages gather the latest tweets from a number of authoritative and reliable sources of government, media and civil society in local languages.
Tiktok's COVID-19 information page has been visited over 52 million times in their five major European markets (UK, Germany, France, Italy and Spain).
During the COVID-19 pandemic, Mozilla has effectively used its browser space (Firefox snippets) and features (Pocket) to promote important public health information from the WHO, leading to more than 35,000,000 impressions and 25,000 clicks in Germany and France alone in the excerpts, while the elaborate coronavirus node in pocket generated more than 800,000 page views from more than 500,000 user uses s all over the world until mid-July. Mozilla also provided expertise and opened datasets on the use of Firefox in February and March to help researchers investigate social distancing measures.
The pandemic is at the centre of the fight against fake news
The reports provide further information illustrating the actions taken to combat COVID-19-related disinformation and the impact of these actions until June 2021. Some examples from the reports as presented by the European Commission:
● Twitter reports that its central COVID hub "covid19.twitter.com" has been visited by over 160 million people, more than two billion times. Since the update last year, in response to the pandemic, they challenged 11.7 million accounts, suspended 1496 accounts (+156 compared to May) and removed over 43010 Tweets worldwide (+5110 compared to May) for covid-19 violation. 19 misleading information policy. Also in June Twitter launched a "101 Course on Getting Started with the Twitter API" for students and researchers conducting academic research to help them better use the Twitter Application Programming Interface Tool (API). The API allows researchers to gain real-time access to large volumes of data from Twitter.
● Microsoft reports that in June 2021, the Bing COVID experience had 2,405,231 views in the EU, showing a 15% increase in views compared to the previous month (+363,692). Microsoft Advertising also prevented 1,583,881 ads that violate its advertising policies (including COVID-19 and vaccines) that touch users in the EU, a significant increase from April (+1,233,427) and a slight increase over March (+55,955).
● Google reported that it took action against 14,895 UrLs in AdSense in June, an increase compared to May (+2,013) and April (+4,346), with the most notable increase in actions taken in France (+1,065 compared to May and +2,370 compared to April). At the same time, the number of ads and ad accounts that were disapproved for violating COVID advertising policies is in line with those reported in April and May.
● Facebook reports that in June it removed more than 76 thousand pieces of CONTENT in the EU on Facebook and Instagram, for violating their COVID-19 policies and for vaccine disinformation, continuing a trend of significant increase from the previous month (+ 14,000). In addition, the company cited its collaboration with researchers from the University of Michigan (MSU) to promote the ability to detect and render fake news.
● TikTok reports on its continued action to implement a COVID tag, with a reduced number of such labels until June in the EU (-26,403). A vaccine tag applies to all COVID-19 vaccine videos, and a banner redirects the user to verifiable, authoritative sources of information. To check the facts there are external experts and inform about the misinformation about the COVID-19 vaccine. Let users know if a video has been marked to contain unsurpassed content. If a user then tries to share the video, they'll be asked to pause and review their next move before they choose to "cancel" it.
