Social Media as News Platform

The Platform Shift

Social media platforms have become the primary news source for younger audiences worldwide. A 2024 Reuters Institute survey found that among 18-24 year olds, TikTok, Instagram, and YouTube were more commonly used for news than any traditional news outlet. This represents a fundamental shift: the gatekeeping function that editors and journalists once performed is now performed by algorithms.

The shift has both democratized and destabilized the news ecosystem. Anyone can now publish and distribute information to a global audience without passing through editorial gatekeepers. Citizen journalists have broken important stories. Marginalized communities can tell their own stories. But the same openness allows misinformation, propaganda, and conspiracy theories to reach audiences just as easily as verified reporting.

Algorithmic Gatekeeping

When algorithms decide what news you see, the incentive structure changes fundamentally. Traditional editors selected news based on importance, impact, and public interest. Algorithms select content based on engagement — what will keep you scrolling, clicking, and watching. Research consistently shows that emotionally arousing content (outrage, fear, amazement) generates more engagement than nuanced, balanced reporting.

Platform companies argue they are neutral hosts, not publishers. But their algorithmic choices — what to amplify, what to suppress, how to rank content — are editorial decisions with enormous consequences. Facebook's algorithm change in 2018 that prioritized 'meaningful social interactions' dramatically boosted political content that generated arguments, inadvertently amplifying polarization. These are not neutral technical choices — they shape the information environment for billions of people.

Editor Logic vs Algorithm Logic

The difference between the two gatekeeping systems is a difference in objective functions:

Worked example: the Facebook Files, 2021

In 2021 the whistleblower Frances Haugen leaked internal Meta research to the Wall Street Journal (the 'Facebook Files') and testified before the US Senate. The documents showed the company's own researchers had concluded that the 2018 'meaningful social interactions' change rewarded reshared, angry content, and that political parties in Europe reported reshaping their messaging to be more negative because that is what the algorithm spread. Crucially, this was not an outside critic's theory — it was Meta's internal data, showing the firm understood the polarizing effect and weighed it against engagement metrics. The case is the clearest evidence that algorithmic ranking is an editorial act with measurable civic consequences, and that 'we are just a neutral host' does not survive contact with the company's own findings.