That distinction matters, inasmuch as for years public discussion of the issue was often reduced to fake news, bot farms, or fact checking and here Kallas is pointing to something broader.
Her argument is that democracy depends on trust, and that once people can no longer tell what is true and what is manipulated, the whole decision-making environment becomes unstable.
In her own phrasing, Foreign Information Manipulation and Interference is “death to democracy by a thousand cuts.”
This is a serious shift in emphasis and the speech suggests that the real target is not simply public opinion on one issue or another. The target is the shared reality that democratic societies need in order to function at all.
If that shared reality breaks down, elections still happen, institutions still stand, and citizens still speak freely, but the information space in which they operate becomes easier to poison.
That is the underlying lesson of the speech, and it should shape how civil society thinks about the problem.
Another important point in Kallas’s remarks is that AI has changed the scale and cost of manipulation.
She contrasts older methods, where states like Russia and China relied on large bot farms and fake profiles, with the current environment in which AI can generate manipulative content quickly, cheaply, and at volume.
Her example is Romania in 2024, where she says a Russia-backed campaign used AI-powered bot accounts, deepfakes, and paid TikTok influencers to artificially boost a fringe pro-Russian candidate.
Her warning is not that technology itself is inherently malign. It is that the economics of interference have become far more favorable to the attacker.
That matters because low-cost manipulation changes the burden on the defender and a lie no longer has to be persuasive in a deep sense.
It only has to be repeated enough, targeted well enough, and timed effectively enough to crowd out trust, create confusion, or distort what feels socially normal.
In that environment, the old idea that truth will simply win if it is available starts to look naive.
One of the most useful parts of the speech is Kallas’s description of a “marketplace for manipulation” where she says interference works because it pays off and describes a world in which actors outsource operations to contractors, private firms, influence-for-hire networks, and digital marketing actors.
She also points to cryptocurrencies, technical service providers, and illicit financial flows as part of the enabling infrastructure.
This is important because it reframes disinformation as something closer to an industry than a collection of random propaganda incidents.
That framing has real consequences because if manipulation is an industry, then the response cannot be limited to media literacy campaigns or post hoc debunks.
It requires financial investigations, sanctions, platform cooperation, cyber inquiries, and regulatory enforcement.
Kallas says plainly that the business model has to be broken and argues that the EU already has tools such as the Digital Services Act, the Code of Conduct on Disinformation, the Rapid Alert System, and the European Cooperation Network on Elections.
Her complaint is not that Europe has no instruments. It is that Europe has not always shown enough political will to use them fully.
That is another lesson worth taking seriously because it shows the weakness is no longer mainly diagnostic. European institutions have spent years studying how these operations work.
Kallas says Europe knows what the weapon is, how it is made, and how it fires. The problem now is whether democracies are prepared to act with enough speed and seriousness before interference scales further.
The Moldova section may be the most important part of the speech for anyone thinking about civic resilience.
Kallas presents Moldova as an example where interference did not win because the response was broader than surveillance or counter messaging alone.
Moldovan authorities publicly warned about the risks, criminalized vote-buying, worked with intelligence services to expose the operation, and benefited from EU support. But she also emphasizes something more social and political.
Moldova ran a major communication effort that highlighted the tangible benefits of European integration, and it did so through a broad coalition that included NGOs, public figures, TV personalities, musicians, and the Church.
She then goes one step further and says that people need to be shown, not merely told, what those benefits mean in practice. In many Moldovan villages, that meant concrete improvements such as parks, playgrounds, and running water.
This may be the deepest insight in the whole speech I think, because she shows that information warfare does not succeed only because hostile actors are clever.
It also succeeds where democratic systems feel distant, abstract, and unable to translate values into lived experience.
If civil society wants to strengthen resilience, then it cannot rely on correction alone.
It has to connect truth, democracy, and Europe to things that people can see, feel, and use in daily life.
A citizen who experiences public benefit directly is harder to detach from reality by pure narrative manipulation.
Kallas also warns that interference campaigns are moving earlier in the cycle.
After what she describes as a failed attempt in Moldova, Russia began targeting Armenia more than a year before its election, and she says similar activity is already visible in Montenegro.
Her broader warning is that every national election is now a target and that 2026 could be a perfect storm in Europe.
That timing point should change how civilians and civic groups think.
If manipulation begins a year in advance, then resilience cannot be built in the final weeks of a campaign. It has to exist beforehand in the form of trusted networks, informed local communities, journalists who understand amplification dynamics, and civic organizations that can explain patterns rather than chase every individual falsehood.
There is also a geopolitical signal in the speech.
Kallas says that over one hundred countries were attacked by FIMI last year and that the withdrawal of US funding for information integrity has weakened global efforts, slowing related work in the G7 and NATO.
Her answer is that Europe must fill the void and treat information integrity as a global public good.
That suggests we are entering a new phase.
Europe is no longer speaking only as a victim of hostile information operations. It is beginning to speak as a potential coordinator of a wider defensive model.
Whether it can actually play that role remains to be seen, but the ambition is clearly there.
For civil society, the implications are straightforward even if the work is not.
The task is not to become a miniature intelligence service or to answer every lie with a fact check. The task is to build trust-rich environments where manipulation has less room to grow.
That means teaching people how amplification works, tracking recurring narratives, connecting institutions to real life, working with local media and community leaders, and preparing long before election season begins.
That conclusion is an interpretation of Kallas’s logic, but it follows directly from the examples and priorities she lays out in the speech.
The most important lesson from Kallas’s speech is that truth on its own is not enough.
It has to be socially carried, institutionally defended, and made visible in everyday life. If hostile actors are trying to manipulate the environment in which democratic judgment happens, then the answer is not simply more content.
The answer is stronger civic ground.
And now, you know.
Source
Kaja Kallas, “Keynote speech by HRVP Kaja Kallas at the 2026 Conference on Countering Foreign Information Manipulation and Interference, ‘From Insight to Impact’,” European External Action Service, 17 March 2026, https://www.eeas.europa.eu/eeas/keynote-speech-hrvp-kaja-kallas-2026-conference-countering-foreign-information-manipulation-and_en
