There is a particular kind of Irish Instagram account that most of us would recognise immediately. Black and white photos of old Dublin streets. GAA county finals from the eighties. A clip of a seisiún in some rural pub, the fiddle player barely visible through a haze of nostalgia. These accounts feel harmless, even comforting. They trade on something genuine: a sense of shared identity, of collective memory, of a place and a people that feel increasingly harder to locate in the pace of modern life.

They are also, in many cases, the top of a very deliberate funnel.

What begins as parish pride and trad music has, for a significant cohort of Irish users, slowly become something else. The same accounts, or accounts that the algorithm surfaced off the back of them, begin introducing new themes. Crime stats with no context. Clips of protests. Posts about fuel prices framed as deliberate government cruelty. Content about immigration described not as policy debate but as cultural extinction. And then, gradually, the logical endpoint: distrust of the media, distrust of elections, calls to action.

This is the Irish Instagram pipeline. It is not a conspiracy. It is a feature.

The feed is not Ireland

Following the fuel protests and the torrent of online reaction that came with them, the sense of a country bitterly divided is hard to shake. But there is an important distinction between what Ireland looks like online, particularly on Instagram and Facebook, and what it looks like when you actually talk to people. The Irish online sphere has its own dominant tones. Instagram and Facebook tend to skew pro-protest, anti-government, and more receptive to conspiratorial narratives. Reddit’s Irish communities tend toward scepticism of the protests and a more liberal outlook. Neither is a representative sample. Most people you encounter offline are more ambivalent, more mixed, or simply disengaged altogether.

The problem is that your feed does not feel like a curated sample. It feels like the world. When the same arguments appear again and again, when every comment section seems to confirm what the last one said, the impression that forms is that this is just what people think now. That there is a “silent majority” that happens to share every view your algorithm has fed you for the past six months. That impression is manufactured, not organic.

I’ve even seen some people sharing personal projects claiming they would share data that the “government doesn’t want you to see”. But when I looked through the data, it shows Ireland isn’t the worst on diesel prices, and that most increases are directly linked to conflict in the Middle East.

Overall, the gap this creates is corrosive. It produces a misperception of majority opinion that makes extreme positions feel mainstream, and it creates social pressure to conform to whatever line appears dominant in your feed. The “silent majority” framing works in both directions: it is how fuel protesters justify their cause, and it is how counter-commenters justify theirs. Both groups are, to a large degree, talking to a mirror.

In the past week, I’ve seen accounts that have never once posted about politics, lifestyle pages, humour accounts, food and travel profiles with hundreds of thousands of followers, have been sharing hot takes on the government and the winding down of the fuel protests. No nuance, no context, just confident opinion delivered to an audience that followed them for entirely different reasons.

Nuance is critical these days, but unfortunately the world now only deals in absolutes.

I’m not suggesting external forces are directing these accounts or paying for placements. What I am saying is that the reach involved dwarfs what many traditional media outlets can claim, and when that reach is suddenly pointed at a politically charged moment, the effect on public sentiment is significant whether it is coordinated or not. The algorithm does not distinguish between a seasoned commentator and a food blogger who decided today was the day to weigh in. It just sees the audience, and it amplifies.

Outrage is a product, not a side effect

None of this is accidental. The architecture of Instagram is designed to maximise engagement, and engagement is not the same thing as truth, value, or democratic health. Posts are ranked heavily on comments and interaction. Divisive content generates more comments than consensus. Rage-bait travels further than nuance. The algorithm does not know or care whether a post is accurate; it only knows that people responded to it.

Crucially, arguing against a post, or even reporting it, often sends a stronger engagement signal than ignoring it. This means that racist or conspiratorial content gets amplified not because it is widely endorsed, but because it generates arguments. The person horrified by the post and the person who agrees with it are, from the platform’s perspective, producing the same valuable commodity.

On TikTok, the failure of content moderation to contain this dynamic is not abstract. Reporting on the platform’s moderation practices, including documented cases of racist, transphobic, and anti-immigration comments passing moderation review, demonstrates how systematically the system fails. What I have noticed on TikTok personally is something subtler but equally deliberate: relatively uncontroversial videos being surfaced with the most inflammatory comments foregrounded, seemingly to provoke a reaction. If you are aware of this mechanism you can, with effort, account for it. Most users are not aware of it, and that asymmetry of understanding is where the real damage accumulates.

There is no equivalent of an editorial meeting. No public service obligation. No separation of opinion from fact that is enforced in any consistent way. Legacy media, with all its genuine flaws, operated inside a framework of accountability, however imperfect. These platforms do not.

It is not teenagers you should be worried about

The public conversation about social media and radicalisation tends to focus on young people. Teenagers on TikTok. Children and screen time. Age limits and parental controls. These are legitimate concerns, but they obscure the group that is arguably more at risk: adults over thirty who have convinced themselves they are too sensible for social media politics, yet whose primary source of information about the world has quietly become their Instagram feed.

This is the user who avoids X because it seems toxic, who does not really use TikTok, who thinks of themselves as a normal, grounded person. But who scrolls Instagram every day. Who has followed a local community page for years and trusts it instinctively. Who watched the fuel protest coverage almost entirely through reels and comment sections. Who has an older relative who was initially pulled into Facebook via wellness content or parish groups, and who now shares posts about “what the media won’t tell you.”

The psychological conditions for this kind of drift are well understood. Media literacy around algorithmic feeds is generally lower in older demographics. Trust in sources that feel local, familiar, or personal is higher. When an account you have followed for three years because of its lovely old photographs of Connacht starts posting about immigration, the established trust transfers. It does not trigger the same scepticism a stranger would.

The consequences extend well beyond what people believe in the abstract. Voting behaviour, protest attendance, and the texture of community relationships are all shaped by content that feels personal and local but is often unverified, anonymously run, or produced entirely outside Ireland.

Who is actually running those accounts

A significant proportion of the Irish-facing content that circulates during moments of political tension does not originate organically. The pattern is visible to anyone who looks: accounts that begin as benign Irish interest pages, building an audience over months or years on nostalgia and community content, before gradually shifting tone toward grievance and political agitation. The shift is slow enough to avoid triggering suspicion but deliberate enough to produce a reliably radicalised audience by the time the pivot is complete.

Beyond that, coordinated posting across multiple pages with identical talking points, slogans, and video clips suggests organisation rather than coincidence. State-linked influence operations, transnational far-right networks, and commercial actors pushing cryptocurrency or financial independence alongside political grievance are all documented participants in this ecosystem in other jurisdictions, and Ireland is not exempt from their attention.

The effect is to impose imported culture-war framing onto Irish issues. Questions about housing, fuel, immigration and crime are genuine and pressing. But the frames used to discuss them online, the vocabulary, the scapegoats, the conspiratorial assumptions, are often borrowed wholesale from British or American online movements and retrofitted for an Irish context. That process of importation erodes trust in Irish institutions, Irish media, and Irish electoral processes, among people who have never engaged with the original source material and have no idea it exists.

The fuel protests and what they revealed

The fuel protests in Ireland exposed something that had been building quietly for a while. The street-level movement had legitimate roots in genuine economic anxiety. But those roots were compromised early, and the compromise was visible in real time for anyone paying attention.

Irish flags began appearing at the protests in a way that should have prompted a hard conversation among genuine participants about who was co-opting their movement and to what end. The experience of the Muslim Sisters of Éire is, in many ways, the clearest illustration of where that failure of vigilance led. The group, which runs a soup kitchen serving homeless people in Dublin, cancelled their service due to the protests. When they offered to facilitate the unhoused alongside protestors, they received racist abuse from people who had threaded themselves into the broader movement. People claiming to speak for ordinary Irish communities turned on an organisation doing practical, unglamorous work to help the most vulnerable people in those communities.

That moment did not emerge from nowhere. It emerged from months of online content that had blurred the line between economic protest and identity politics, between genuine grievance and manufactured contempt. The algorithm built the conditions; the street was where they found expression.

Regulation that treats this seriously

Ireland’s regulatory response to social media has, so far, been calibrated almost entirely to the child safety conversation: age limits, screen time, online safety. These are not wrong concerns. They are simply not the only ones, and focusing on them exclusively allows a much larger problem to remain unaddressed.

There is no requirement for large Irish-facing political accounts to disclose their funding, their operators, or their country of origin. Irish users have no meaningful way to understand why they are being shown certain political content, or who decided they should see it. The platforms are treated as technology companies with a youth safety problem, not as political infrastructure with a democracy problem.

A more serious response would start from a different premise. When a platform functions as the primary news source for a substantial portion of the adult population, it is performing a public function regardless of how it is legally classified. That should come with obligations: transparency for accounts that operate at scale, labelling of algorithmically curated political content, access for independent Irish researchers to platform data on how these dynamics are playing out here rather than extrapolating from American and British studies. It would also mean pushing at EU level for stronger enforcement of existing obligations and new ones where gaps are visible.

What individuals can do in the meantime

Telling people to delete Instagram is not a useful answer. Most will not, and for many the platform is how they maintain genuine social connections. What is useful is a shift in how people relate to what they see there.

The most practical starting point is recognising that your feed is not Ireland. It is a slice of Ireland that has been assembled by an algorithm optimised for engagement, not accuracy or balance. Deliberately following a wider range of sources, including ones you find irritating, is one corrective. Being sceptical of accounts that pair warm nostalgia with increasingly simplified and angry political content is another. Treating viral clips of protests or violent incidents as the beginning of a story rather than the whole of it is a habit that takes effort but is worth building.

At a national level, media literacy education needs to extend beyond children and beyond traditional media literacy into something that addresses how algorithmic feeds specifically work, what influence operations look like in practice, and how to recognise when content that feels local and personal is actually neither.

The Irish internet and real Ireland are not the same place. We would do well to remember that before the gap between them gets any wider.