---
id: recommender-systems
name: Recommender Systems
family: technological
tier: active
url: https://explorer.talin.digital/e/recommender-systems/
author: Benjamin Talin
citation: Talin, B. (2026). The Future Explorer. https://explorer.talin.digital (dataset v0.4.0, 2026-11-16)
license: CC-BY-NC-4.0
---

# Recommender Systems

The software that decides what you see next on a feed, video site or shop, ranked by what is most likely to keep you watching rather than by what you asked for. On one big video platform it chooses over 70% of what people watch. When researchers reduced how many like-minded posts thousands of users saw, their political views did not measurably change. It shapes what billions of people see, and so what seems normal.

> Software that decides what you see next, ranked by what is likely to hold your attention rather than by what you asked for.

## Maturity
5/5 — Ubiquitous and mature: it decides most of what billions of people see each day. The EU now requires a non-profiled alternative; people still use the ranked feed. Independent measurement of its effects is the thing that is missing.

Capability 5 · deployment 5 · absorption 5 · bound by regulation · flat

Source: Meta, Q4 2023 results (1 Feb 2024); Neal Mohan, CES (Jan 2018), via NBC News; Regulation (EU) 2022/2065, Art. 38; European Commission (25 Apr 2023); Guess et al., Science 381 (Jul 2023); Nyhan et al., Nature 620 (27 Jul 2023) (2018–2024)

## What it is
Algorithms that decide what a person sees next, optimised against engagement signals rather than against stated preference.

## What it does
It is the mechanism by which attention is allocated at population scale. It shapes what is seen, and therefore what is believed to be normal - which is a claim about social reality, not about software.

## Common misconception
That it reflects preference. It optimises a proxy for engagement, and the gap between what people engage with and what they would choose on reflection is where the harm sits. The effect on polarisation is genuinely contested and must be presented as such.

## Steepens
- The Attention Economy (the-attention-economy) — large
- Platform Mediation of Social Life (platform-mediation-of-social-life) — large
- Epistemic Fragmentation (epistemic-fragmentation) — large
- Polarisation and Radicalisation (polarisation-and-radicalisation) — moderate
- Time Migration into Digital Environments (time-migration-into-digital-environments) — moderate

## What would change its gradient
- Researcher data access mandated rather than negotiated
- Interface defaults - chronological or user-set ranking as an option that persists
- Liability attaching to amplification rather than to hosting
