Recommendation Systems
Algorithms that automatically select and rank content to show each user, typically optimized to maximize engagement rather than quality or accuracy.
Automated systems that filter, rank, and personalize content for individual users based on their behavior, preferences, and data. Most social media and streaming platforms use them to decide what to display in feeds, search results, or homepages. They are typically optimized for engagement metrics like clicks, watch time, or shares rather than user wellbeing or information accuracy.
What this means in real life
When you watch one cooking video on YouTube, the platform's system learns your interest and begins suggesting similar recipes and cooking channels—showing you content it predicts you'll watch longest, not necessarily what's most nutritious or accurate.
What it isn’t
Not simply a neutral sorting tool or search function. Many assume these systems just organize existing content by relevance; in reality, they actively shape what billions of people see by amplifying certain content and burying others based on engagement predictions.
Commonly misused online
Often used interchangeably with 'algorithm' broadly, when it specifically refers to personalized content-ranking systems. People tweet 'the algorithm showed me this' to mean any unexplained online discovery, conflating recommendation systems with all algorithmic decision-making.