DUMONDE
Essay 4 / 7Working Paper19 min

The Algorithm of Polarisation: How Platforms Amplify Division

|Updated February 4, 2026|Introductory
In 2021, Frances Haugen, a Facebook engineer, leaked internal documents that revealed something the company already knew: its algorithm optimised for division. The company's own researchers had documented it. It was not a bug. It was the product working as designed.
algorithmspolarisationfacebookyoutubedisinformationhaugen

Abstract

In 2021, Frances Haugen revealed that Facebook's algorithm optimised for division. This essay analyses how recommendation algorithms amplify divisive content, why false news travels faster, and YouTube's rabbit holes.

Cite this work

Discussion

Methodology

What are the strengths and limitations of this approach?

No comments yet. Be the first to share your thoughts.

Implications

What practical implications does this research suggest?

No comments yet. Be the first to share your thoughts.

Extensions

What future directions would you explore from here?

No comments yet. Be the first to share your thoughts.

Comments

No comments yet. Be the first to share your thoughts.

Sign in to join the discussion