Hungary just ran the most expensive experiment in media capture in EU history. The results were not what the experimenters hoped for. From the Media Finance Monitor.
I spent election night at a watch party where a small but vocal faction (I’ll admit I was among them) kept lobbying to switch over to the pro-government channels. The programming there was, let’s say, instructive. People who had spent 16 years telling Hungarians what counted as reality were now receiving some new information on the subject. It was not a gracious evening for them. I am trying to be a bigger person about this, though I am not fully succeeding.
Tisza had the broadest electoral victory in the brief history of Hungarian democracy. There are many reasons why that happened: economic mismanagement, corruption, arrogance, the moral collapse after the clemency scandal, the general decay that sets in when a political system stops receiving honest feedback, and a broader international dynamic in which voters seem increasingly willing to punish incumbents. But most of you are not here for my comprehensive theory of Hungarian electoral behavior.
I do think part of the story has to do with the structure of the information ecosystem and how that structure changes the value of media capture. That seems worth exploring here because it is, at heart, a media-and-money question, which is the only kind we [at the Media Finance Monitor] pretend to have authority over.
It is about how political actors convert resources into attention, and attention into influence and power. It is about what happens when a governing party spends 15 years building an enormous captured media system, only to discover that the information environment underneath it is shifting in ways that make that investment less effective than it used to be.
So I want to spend this edition on that narrower point: not on every reason Orban lost, and not on what Tisza’s victory will mean for the Hungarian media landscape, which we already touched on last week, but on why structural changes in the information ecosystem may have made media capture a weaker political technology than it once was.
Viktor Orban’s Empire of Attention
Over the past 16 years, Fidesz built a remarkably sophisticated machine for dominating the information ecosystem, and it did so in roughly three layers.
The first layer was interference on the advertising market. This is the part we have covered before, with extensive evidence for both the scale of it and the damage it caused. The Hungarian state remained one of the largest advertisers in the country and consistently directed spending toward loyal outlets while starving independent – or merely insufficiently obedient – ones. Most private advertisers, especially those operating in heavily regulated sectors, drew the obvious conclusion and followed suit.
The result was a media market in which it became extremely difficult to run a non-aligned business sustainably. Weakening outlets economically made them easier to intimidate, marginalize, and, eventually, to acquire by government-friendly businessmen.
The second layer was capture and consolidation. Once enough of the market had been distorted, ownership could be consolidated into an enormous pro-government portfolio of newspapers, radio stations, television channels, online outlets, and local titles. By the late 2010s, the government and its allies had reached critical institutional mass through the Central European Press and Media Foundation and other properties. This mattered especially outside Budapest, where old consumption patterns held longer and where the offline dominance of aligned media remained particularly strong.
The third layer was adaptation to the platform age. And here, in a coldly technical sense, one has to acknowledge the innovation. Fidesz understood earlier than many of its critics that owning newspapers, radio stations and television studios, while still useful, was no longer enough in a platform-mediated information environment. Attention was moving elsewhere, into platforms that were too big and too unruly to capture with the same methods they had used so effectively against Hungarian media.
The governing camp did not abandon its captured institutional empire; it built a second system on top of it. This newer system crystallized mostly around the Megafon Incubator Center that relied on platform-native voices, pro-government influencers, and heavily amplified, low-grade political content designed less to persuade than to saturate. Censorship by noise, if you will.
The point of this model was not primarily to win elegant intellectual arguments. Much of the content was too crude, too aggressive, and too transparently instrumental for that. Its function was to flood newsfeeds, reward loyal messengers, smear opponents, intimidate critics, and make participation in public discourse feel exhausting, ugly, and sometimes personally risky. If this was the discourse on offer, many sane people would simply decide they wanted less of it.
They spent enormous sums of money amplifying these voices. In the first nine months of 2025 alone, government-affiliated actors accounted for 87 percent of the 4.1 billion forint (10.6 million euros) spent on Google and Meta ads in Hungary. That is an extraordinary resource advantage, and it helps explain why this strategy worked for as long as it did. The ruling camp was able to turn money into platform reach at a scale no opponents could even begin to match.
Fidesz built a system that could shape institutional distribution, dominate offline visibility, and carpet-bomb platform feeds with aligned messages. It promoted government narratives, distorted public discourse, and raised the cost of dissent. For a while, this machine worked exceptionally well.
Ad Ban Jams the Information Autocracy
The institutional layer was already in trouble before any of this came to a head. The newspapers, radio stations, and television channels that formed the backbone of the captured empire were losing audiences the same way legacy media everywhere was losing them. The Megafon model was itself an acknowledgment of this: you don’t build an expensive parallel system of platform-native influencers if your existing portfolio is doing the job. The empire adapted because it had to.
But Megafon had a structural problem that money was papering over. The content was low quality, the voices largely unconvincing to anyone not already converted, and the organic reach was minimal. This is precisely why the amplification spending was so enormous. The 87 percent market share of political ad spending was the cost of keeping an artificially inflated system running. Without the money, the reach collapsed.
Then came October 2025, and Meta’s decision to exit political advertising in the EU entirely, with Google following on a similar timeline. Spending that had run to billions of forints in the pre-ban period dropped to a fraction of that in traceable form, with what remained pushed into proxy advertisers, misclassified ads, and workarounds that were slower, less effective, and harder to scale.
The machine that had been able to flood feeds on demand no longer had the fuel to do it, and Fidesz’s ability to shape public discourse diminished significantly.
Decentralized Information Resists Capture
There is a standard lament in journalism and media circles about the shift from institutional media toward creators, influencers, and platform-native voices. It goes roughly like this: institutions have editorial standards, verification processes, legal accountability, and professional norms that took decades to build. Individual creators have none of these things, or have them only loosely and selectively. As audiences migrate away from institutions and toward individuals, a lot of these benefits are lost, and the information ecosystem becomes noisier, less reliable, and more susceptible to manipulation.
A lot of this is true. A person on YouTube with a ring light and strong opinions is not a straightforward replacement for a properly resourced investigative newsroom. The hollowing out of institutional journalism has real costs, and we have written about them here at length.
But Hungary has just run a fairly compelling live experiment in favor of the other side of the argument: decentralized information ecosystems are structurally harder to capture.
You can buy a newspaper, and you can starve a broadcaster of advertising revenue until its owners become negotiable. You can hire a hundred influencers, put them on state payroll, and spend billions of forints amplifying their content across every platform available to you.
What you cannot do, at least not as easily or as cheaply, is buy the entire information environment when it is distributed across thousands of voices with their own audiences, their own revenue streams, and no single point of institutional control. The very features that make the creator economy alarming to many people in legacy journalism – its lack of central coordination, its weak attachment to editorial protocol, its organizational informality – are also what make it resistant to the kind of top-down capture the Hungarian model depended on.
This is not a universal argument. In harder autocracies, governments that face a decentralized information ecosystem they cannot capture through financial means often go after the infrastructure instead: throttling platforms, blocking applications, firewalling or turning off the internet entirely, criminalizing access to independent information, or making dissent directly dangerous. Where leaders are willing and able to do that, decentralization offers much weaker protection.
But in competitive information autocracies, like Hungary, where power distorts the media heavily while stopping short of switching off the lights entirely, the dynamics are different. More than 3 million votes for Tisza suggests that in such systems, the ceiling on what capture can achieve may be lower than it once appeared.
(And watching pro-government television on election night was not, as some have suggested, schadenfreude. It was primary source research to prove this point.)
Peter Erdelyi is the founding director of the Center for Sustainable Media, where this article originally appeared in the Media Finance Monitor newsletter. Republished by permission.
