Social media giants TikTok and Meta faced a fierce competition in the algorithm arms race. Whistleblowers revealed that these companies prioritized engagement over user safety. An engineer at Meta admitted allowing more ‘borderline’ harmful content, such as misogyny and conspiracy theories, in user feeds to compete with TikTok. TikTok’s internal dashboards showed that staff were instructed to prioritize cases involving politicians over harmful posts featuring children. A senior Meta researcher, Matt Motyl, stated that Instagram Reels, launched in 2020, lacked sufficient safeguards. He noted that Reels posts had a higher prevalence of bullying, hate speech, and violence compared to other content on Instagram. These revelations highlight the risk companies take to maintain a ‘strong relationship’ with political figures to avoid regulations or bans, rather than prioritizing user safety.
Meta launched Reels without adequate safety measures, despite needing 700 staff for growth and refusing to hire specialists to protect children and ensure election integrity. Research documents show that Facebook recognized issues with its algorithm, but the incentives of the current set-up do not align with their mission. Engineers relied on content safety teams to handle harmful posts, but the rapid changes in algorithms risked increasing harmful content. Teenagers reported receiving violent and hateful content, and a young man named Calum described being radicalized by the algorithm.