At virtual shareholder meeting, Mark Zuckerberg says that Facebook's criteria for policing content should remain limited to avoiding “imminent harm”
Jill Goldsmith / Deadline :
Context & Ripple Effects
This shareholder-meeting answer extends a line Zuckerberg has run for years: scope moderation as narrowly as possible and resist broad content categories. He made the same move on deepfakes in 2019, when he argued the definition was too broad and needed tighter scoping rather than expanded.
The position also previews the legal architecture he would later propose explicitly: [[a:964477|conditional Section 230 protections tied to platforms' ability to identify and remove content]]. If liability hinges on removal capability while policy only targets imminent harm, the platform itself sets how much gets policed.
First-order effects
- Shareholders pressing Facebook on content policy get the company's operating threshold on record: enforcement limited to imminent harm, with everything else left up by default.
- Advertisers and civil-society critics now have a concrete criterion to litigate against — any content causing non-imminent harm falls outside Facebook's stated policing remit.
Second-order effects
- Lawmakers weighing Section 230 changes must reckon with a framework where the same company arguing for conditional protections defines the harm that triggers them — the regulator's baseline and the platform's policy become the same document written twice.
- Rival platforms face pressure to publish their own harm thresholds, since Facebook has staked out the permissive end of the spectrum and can frame stricter peers as over-moderating.
Third-order effects
- If imminent-harm scoping holds as both policy and the basis of immunity reform, content governance structurally shifts toward distribution-layer liability: rules attach to what platforms remove, not what they host, giving incumbents durable control over where the line sits.
- The pattern points toward regulation built around each platform's self-declared enforcement capacity — an incentive structure that rewards under-scoping harm definitions rather than expanding them.
The trend: Content moderation is converging on a model where platforms themselves define the harm threshold that both their policies and their legal protections rest on.