"Social media" means nothing — and that matters more than the platforms realize

On 10 July 2026, researchers at Aston University published a finding that should embarrass every legislator who has drafted a "social media" bill in the last decade: ordinary people cannot agree on what the term means, and they cannot reliably say which websites fall inside it.

The study, reported that morning across the UK press, surveyed a representative sample of adults and asked them a question the platforms, the regulators and the public have all been pretending is settled. It is not. Two people sitting in the same room will give two incompatible lists. A regulator drafting age-verification rules, a platform writing its terms of service, and a parent trying to police a teenager's screen time are operating with no shared map of the territory.

A term that has eaten its own meaning

"Social media" worked, as a phrase, for about fifteen years — long enough to outlive the products it was invented to describe. Friendster, MySpace, even the early Facebook were recognisably the same kind of thing: a profile, a list of friends, a feed. The label held.

It does not hold any longer. The Aston team found that respondents routinely included or excluded YouTube, WhatsApp, TikTok, Discord, Reddit, LinkedIn, Wikipedia's talk pages, X, Threads, Bluesky, Snapchat, dating apps, group iMessage threads, and even some gaming platforms, depending on how they read the question. The boundaries are not just fuzzy at the edges — they disagree about the centre. A messaging app with a billion users is sometimes counted, sometimes not. A video platform with no friends list is sometimes counted, sometimes not. The only consistent signal is that the public has no consistent signal.

That matters for reasons that have nothing to do with semantics. If "social media" cannot be defined, then every law, every compliance regime, every academic study, every advertising dollar, every parent's rule that uses the phrase as its subject is operating on undefined grounds.

Who benefits from the fog

The platforms benefit, obviously. "Social media" is the bucket into which the public drops every unpleasant online experience — addictive feeds, harassment, misinformation, teen mental-health harms, mass surveillance. When the bucket cannot be defined, the burden of definition falls on the complainant, the regulator or the parent. The defendant can always say: we are not that.

YouTube has run this play for years. Its parent company Alphabet has argued, in regulatory filings in the United States, the United Kingdom and the European Union, that YouTube is a video platform with social features, not a social media platform proper — and therefore should be exempt from rules written for social media. The argument is not entirely frivolous, but it is also not something users accept when asked. It is a corporate taxonomy imposed on a public conversation.

Regulators benefit, more quietly, because undefined terms give them discretion. A "social media regulator" whose remit extends only to "social media" can claim jurisdiction over the apps it wants to regulate and decline it over the apps it does not. The phrase does regulatory work that no precise statute would allow.

Researchers benefit as well, though they will not say so. The literature on "social media and mental health," "social media and democracy," "social media and elections" runs into the tens of thousands of papers and assumes a stable object of study. It is the methodological equivalent of measuring a coastline without agreeing which instrument to use.

The structural frame

What the Aston finding exposes, in plain terms, is that an entire regulatory and academic ecosystem has been built on a category error. The category looked stable because the largest platforms — Facebook, then Instagram, then Twitter, then TikTok — were obvious members of it. Once the obvious members start to fragment, and once adjacent platforms (video, messaging, gaming, dating) start to absorb the social functions, the category becomes a corporate slogan rather than an analytic category.

This is the same pattern that has played out in every platform-adjacent industry: a hot, fast-growing sector is named for what it does, the name hardens into a regulatory object, then the sector diversifies faster than the regulation. "Search" survived because Google kept the meaning narrow. "Social media" did not survive because no one firm held it. The phrase has become a folk category — useful in casual speech, worthless in legislation.

Stakes, and what to watch

The practical question, now, is whether the platforms, the regulators and the researchers will move to a more honest vocabulary, or whether the fog will harden into permanent ambiguity. Several concrete things follow.

First, the United Kingdom's Online Safety Act, the European Union's Digital Services Act, and the United States' various state-level age-verification laws all use "social media" or close cognates as a scoping term. If that term is contested in the public mind, every compliance filing under those regimes is now contestable on definitional grounds alone.

Second, age-verification mandates — the most active area of UK and US regulation in 2026 — depend on the platforms themselves to identify whether they are in scope. Under-defined categories give the regulated party the call. That is the worst possible place for the line to sit.

Third, the academic literature will struggle to compare across studies if "social media" denotes different objects in each paper. Meta-analyses aggregating dozens of trials on teen wellbeing will, increasingly, be aggregating apples and gaming consoles.

The Aston study will not, on its own, change any of this. But it gives every careful legislator, careful regulator and careful journalist a citable basis to demand a more precise vocabulary. The next draft of any platform bill should not contain the phrase "social media" without a statutory definition. The next parent–teen screen-time negotiation should probably start there too.

This publication framed the finding as a structural problem of platform governance rather than a quirky survey result, because the legislative and academic consequences are concrete and immediate. Monexus will watch the UK Department for Science, Innovation and Technology's response and any subsequent Online Safety Act amendments for signs that the definitional gap is being closed — or quietly left open.