Meta faces scrutiny as hate‑content removals drop 79% in Australia

Australia’s Royal Commission on Antisemitism and Social Cohesion, established after the December 2025 Bondi Beach terrorist attack that killed 15 people, heard testimony that Meta’s policy changes in early 2025 led to a sharp decline in hateful content removals. Between July and September 2025, Facebook actioned 1.2 million pieces of hateful conduct – a 79 % drop from earlier levels – and Instagram saw a 73 % drop in proactive removals.

Meta director Benjamin Good explained that the 2025 reforms ended third‑party fact‑checking, shifted to a Community Notes model, lifted certain topic restrictions and revised the Hateful Conduct policy, removing protections for LGBTQ + users, especially trans and non‑binary people. The changes were presented as a way to “allow more content, more speech” and reduce over‑enforcement, relying more on user reports. Critics including GLAAD and the Human Rights Campaign warn that the revisions have made the platforms unsafe for LGBTQ communities, giving Facebook a safety score of 40 and Instagram 41 out of 100. The commission is questioning whether the drop in removals leaves harmful material online.