Royal Commission Grills Meta’s New Gamble

Meta’s moderation shift is now under fire in Australia because the company admits it will catch less harmful content while taking down fewer innocent posts.

Quick Take

Meta told the Australian Royal Commission that its new model uses more user reports and less proactive scanning.
Commission evidence pointed to a sharp drop in hateful conduct removals after the policy change.
Meta says the change is meant to reduce mistaken takedowns of lawful speech.
The dispute now centers on whether fewer removals mean better accuracy, or weaker protection.

What the Commission is pressing Meta to explain

The Australian Royal Commission is pushing Meta to justify a major change in how it handles hate speech and other harmful content. Commission testimony and reporting say Meta reduced proactive enforcement, raised the bar for automated action, and now waits for users to report some content before acting. Meta says that shift lowers errors, but critics say it also leaves more harmful material online.

That tension sits at the center of the hearing. The company’s policy lead, Benjamin Good, said Meta’s move away from proactive moderation was designed to reduce mistaken removals. He also confirmed that the drop in hateful conduct removals cited by the commission was “roughly correct,” while saying he could not tell whether the decline was caused only by the policy change.

Why the removal numbers matter

Commissioner Virginia Bell cited Meta data showing hateful conduct removals fell from 5.8 million in October through December 2024 to 1.2 million in July through September 2025. That is a steep decline, and it gives the Commission a concrete figure to press in public. But the data on its own does not prove that antisemitic content increased, only that Meta removed far less of it under the newer approach.

Meta’s own explanation is narrower. The company says it is trying to stop over-enforcement and catch fewer innocent users in its systems. In a January 2025 statement, Mark Zuckerberg said the change would mean Meta would capture fewer harmful posts while also reducing wrongful takedowns. That trade-off is now the heart of the fight: accuracy for some users versus more exposure to abuse for others.

Offensive language and real-world failures

The commission also highlighted internal examples suggesting Meta now allows some phrases that previously would have triggered action. Reporting from the hearing said documents showed phrases such as “immigrants are scum,” “white people are all Nazis,” and “gay people are sinners” may now pass under the revised standard. Supporters of the change call that a move away from censorship. Critics see a line that has been pushed too far.

In 2024, Meta announced that it would crack down on "Zionist" used as coded language for Jew-hatred, formally adopting the IHRA definition of antisemitism across Facebook, Instagram, Threads and WhatsApp.

This past week, Meta's director of content policy, Benjamin Good,… pic.twitter.com/mKM7Z00Bpz

— Trisha Posner (@trishaposner) July 7, 2026

Older enforcement failures have added weight to that concern. Meta admitted that a Holocaust denial post stayed on Instagram from 2020 to 2023 even after six reports, and only two of those reports were reviewed by human moderators. Separate hearing material also said four antisemitic posts appeared during the commission process and were not removed at first. Those examples feed a broader public worry that slow review can fail badly when hate spreads fast.

The bigger political and regulatory fight

This dispute is bigger than one company or one hearing. It reflects a wider battle over who sets the rules for speech online, and how much power platforms should have over public debate. Meta argues that reactive moderation reduces mistakes and gives users more room to speak. The Commission argues that a softer system can let dangerous content stay visible long enough to matter, especially when reports come in late.

The practical problem is enforcement. Royal Commission findings are advisory, so Meta does not face an immediate legal penalty from this hearing alone. That leaves the company with room to defend its model while lawmakers, regulators, and advocacy groups keep pressure on it. For readers on both the left and the right, the larger issue is familiar: a huge private platform is making decisions that shape speech, trust, and public safety, with limited public control.