Sources: Facebook letting go of entire Trending Topics editorial staff, consisting of 15-18 workers contracted through a third party
Facebook will no longer employ humans to write descriptions for items in its Trending section, which attracted controversy over allegations of political bias in May.
Context & Ripple Effects
The May controversy set this up: current and former workers told the New York Times the Trending Topics group was ill-managed and leaned on personal news judgment, feeding allegations of political bias even as they denied any institutional slant (that reporting framed the team as both sloppy and scapegoated). Facebook's answer three months later is to remove the humans entirely.
Dismissing all 15-18 contracted editors through its third-party staffing arrangement, Facebook is simultaneously converting Trending to automatically selected topics and excerpts — replacing written story descriptions with bare topic labels. It matters because Trending was one of the few Facebook surfaces where editorial judgment visibly shaped what users saw.
First-order effects
- All 15-18 Trending Topics contractors lose their roles at once, since the third-party staffing model means there are no internal transfers — the entire human curation function is eliminated, not relocated.
- Publishers lose the editorially written descriptions that gave Trending items context, competing instead for placement in an automatically assembled list of bare topics.
Second-order effects
- With no human filter between trending signals and what users see, the surface becomes fully dependent on algorithmic detection of what is spiking — raising the stakes for whoever can move those signals, including low-quality publishers optimized for engagement rather than accuracy.
Third-order effects
- The pattern holds through to the endgame: within two years Facebook shut Trending down altogether (replacing it with publisher-applied breaking-news labels), and along the way stripped personalization from it (showing regional rather than individualized stories) — a curated news module squeezed out first by politics, then by economics.
- Structurally, this was an early instance of platforms retreating from visible human editorial judgment on news surfaces, leaving distribution decisions to ranking systems whose choices are harder to audit or litigate than an editor's.
The trend: Social platforms are replacing visible human news curation with algorithmic selection under political pressure — a retreat that here ended not in better automation but in the product's elimination.