European Commission report: IT companies, including Facebook, Google, Twitter, and Microsoft only review 40% of material flagged as hate speech within 24 hours
Mark Scott / New York Times : See also Mediagazer
Context & Ripple Effects
Six months after Facebook, Twitter, Google, and Microsoft signed the voluntary EU code of conduct on hate speech in May 2016, the Commission's first monitoring round delivers a poor grade: only 40% of flagged material gets reviewed within 24 hours, and just 28% is removed. The report matters because it establishes the baseline for a measurement regime the Commission intends to repeat.
The four signatories are now on notice that compliance will be tracked publicly and by name — a setup that turns each subsequent report into either vindication or embarrassment for the individual platforms.
First-order effects
- Facebook, Twitter, Google, and Microsoft face public, company-by-company scoring on a metric they agreed to — with the Commission signaling it will keep publishing the numbers rather than legislating yet.
Second-order effects
- The shaming dynamic works fast in the coverage arc: removal rates climb to 59% within six months (the mid-2017 follow-up) and reach 81% reviewed within 24 hours by early 2018, when Instagram and Google+ join the initiative — evidence that the reporting itself pulls new platforms into compliance.
Third-order effects
- Voluntary scorekeeping hardens into standing infrastructure: by 2019 firms assess 89% of flagged content within 24 hours, though the 2021 report shows removals slipping back to 62.5% — suggesting the gains persist only as long as the Commission keeps auditing, which is precisely the argument for converting soft commitments into binding law.
The trend: EU content-moderation policy is evolving from voluntary platform pledges toward permanent, quantified Commission oversight — with each monitoring report raising the floor and building the case for enforceable rules.