Twitter has provided an outline of its approach to tackling disinformation and misinformation in Australia, but it has stopped disclosing detailed data regarding the removal of content and the sharing of customer data with authorities.
In a report addressing its compliance with a voluntary industry code for combating false and harmful content, Twitter defended its “new products and policies” designed to protect user rights and safety. However, unlike other signatories to the Australian Code of Practice on Misinformation and Disinformation, Twitter did not include statistics on enforcement actions in its report.
While Meta reported removing 91,000 pieces of content on Facebook and Instagram in Australia for violating harmful health misinformation policies, and Microsoft’s Bing took defensive measures in response to 45,100 Ukraine-related queries, Twitter has chosen to omit such enforcement actions from its industry code report. This decision follows Twitter’s announcement in April to discontinue publishing biannual transparency reports containing this data.
Unlike Google, Meta, and TikTok, which released transparency reports disclosing compliance with Australian government requests for customer data disclosure and content removal, Twitter’s report did not provide similar transparency.
The report stated that Twitter underwent transformational change in Q4 2022, making some examples and data unavailable or less relevant. The report mentioned new approaches, including community notes, the Twitter Blue subscription, and open-sourced algorithms, aimed at moving beyond a binary approach of leaving content up or taking it down.
However, these approaches are not considered sufficient under Australia’s content moderation regulations, such as the Online Safety Act 2021 and the Telecommunications Act 1997, which require companies to block or remove content deemed harmful by agencies.
Twitter acknowledged that Australian users were contributing annotations to misleading tweets through community notes. If a sufficient number of contributors from different perspectives rate a note as helpful, it will be publicly displayed on the tweet. Several high-profile corrections have been made in Australia through community notes, exposing inaccurate information posted by public figures.
Twitter’s $13-per-month Twitter Blue subscription, which amplifies subscribers’ tweets, was mentioned as a measure to combat disinformation. However, doubts were raised about its effectiveness, with concerns that it allows the purchase of influence over political discourse without proper checks and balances.
Additionally, Twitter highlighted its open-sourcing of sections of its content ranking algorithm, fulfilling the industry code’s requirement to provide users with access to information about its recommender systems and options related to suggested content.
Overall, while Twitter outlined its efforts to address disinformation and misinformation, the absence of detailed enforcement action data raises concerns about transparency and accountability in combatting harmful content in Australia.
