Special committee on the European Democracy Shield Ordinary meeting
Legend
Transcript markers
- text Sentence being played
- text Search result
- text Intervention cited by the link you opened
Click a sentence to play the video from that moment. Select text to quote it with the speaker, timestamp and link.
More options
Offset
Transcript offset
Export
Live capture
Versions of this video
Transcripts may differ slightly from one version to another.
EUDS European Parliament
Follow its sittingsGeorgios Giannopoulos
Thank you very much, Mr. Chairman, for this introduction. Good morning also from my side. As you heard, I'm Georgios Giannopoulos, acting director of the Directorate E, Societal Resilience and Security, at the Joint Research Centre of the European Commission. And in this directorate, we are working on several aspects that cover the domain of security. I heard before mentioning the hybrid threats, and this is something that we have been working a lot.
Indeed, it's a very encompassing concept, and of course, drones is one of these threat vectors that can be part of a hybrid campaign. The work that we have been doing at the JRC is not something new. It's something that we have started more than 10 years ago when we realized that this disruptive technology could have serious security implications.
And what we have seen as weak signals a decade ago actually became today a major security concern. So in the meantime, before reaching this maturity level that we have today, we have worked on several aspects related to drones and we have developed the guidelines that you just heard for the protection of public spaces and critical infrastructure against drones. We have produced guidelines to support operators and public authorities to design protection measures against drones. We have been doing testing. We have set up a database in order to register incidents that are taking place across Europe.
So we have done a comprehensive set of activities that led to the establishment of the Counter-UAS Living Lab. We are in a unique position in the JRC having all these secure sites, Ispra, Geel, a nuclear site, where we have already a very good background in terms of security and this enables us to have this site as living labs where we can test certain technologies. And of course we are bringing this concept now even forward with the Center of Excellence and I will come to this in a minute. Maybe next slide, please. Very good.
Of course we are all our activities is very well grounded on policy actions. There is the action plan on drone and counter-drone security which came out in 2026 and there it is explicitly mentioned that the EU will set up a counter-drone center of excellence. The plan is to establish this by Q1 2027. So the center of excellence will launch testing programs, integrating of course aviation safety requirements. And the idea is to develop harmonized testing methodologies for countering unmanned aircraft systems.
Next slide please. Now, what is important to understand is that drones is a very complex technology to a certain extent. It covers air drones, unmanned aircraft systems, but also underwater, land and surface drones. So it's not only the unmanned aircraft systems, although these are the ones that monopolize to a certain extent the interest in the media.
Nevertheless, we had incidents of overwater actually drones that have created some security concerns. For the time being, for the short-term priority of the center of excellence, we will focus on unmanned aircraft systems, excluding HAPS, which are the high-altitude platform systems like balloons, for various reasons. So these are not included in our effort. And the long-term priority, what we have in mind is to expand the capabilities of the center of excellence so that we can work also on underwater, land and surface drones. But this most probably will be part of our efforts in the next MFF.
And we think that the long-term, if you want, existence and performance of the center of excellence is very much linked with the discussions on the next MFF. For the unmanned aircraft systems, we go from the open category, you know, small drones that you can find in the market, up to certified larger-scale drones for which we need also to take into account aviation safety requirements. Next slide, please. Our main site and infrastructure of this living lab and soon-to-become center of excellence is in our site in Geel.
Geel is I think an hour or something from here, from Brussels. It's a nuclear site and as I said, this is an evolution now that we are going through from the living lab to the Center of Excellence. And what this will include, and we mention here main site and infrastructure because the Center of Excellence is something broader, as you will see in the next slides. So what we will have there is research hub to do independent scientific research. This is really at the core of our mandate at the JRC, to perform independent scientific research which can feed in the policy cycle of the European Commission.
So this is our bread and butter, if you want, in the JRC. We go one step further, we want to make use of our knowledge but also our networks to support practical training for Member States' law enforcement. So it will be also kind of a training center, if you want, in the near future. Of course, one of the most important activities of this will be the testing of counter-drone systems, so live testing of detection, identification, and tracking systems. And here I need to make one very important point.
We don't work on mitigation measures. Mitigation measures means, to put it very bluntly, bringing down drones when they enter in a restricted area. This is something that we will not test this kind of systems. For the time being, we don't have this capability in the Geel site. So our activities will be limited to detection, tracking, and identification.
The mitigation is something that we cannot do in this center of excellence, but we are expecting in the Geel site, but we are expecting that this can be done in collaboration with Member States by using their satellite centers. And of course, as I mentioned, policy support activities, producing handbooks like the ones that we have already produced, technical reports, and policy guidance. This is again very much in line with our core business. Next slide, please.
As I said, the concept of the center of excellence is not just to have one site in Geel doing this kind of activities, but to connect also with national sites that complement what we cannot do for various reasons. Each satellite center or site would be tailored to tackle specific counter-UAS scenarios, for example maybe more on detection or maybe more on tracking or maybe more on mitigation, which is also something that we will not cover in Geel. So for the time being we are considering various satellite centers, as you can see here on Southern Europe, Western Europe, Northern Europe, Eastern and Central Europe. That is because we want to cover also different threat scenarios. You know, you have different needs also depending on which geographical area you are.
The needs and the specificities of the Mediterranean, for example, is different compared to the eastern borders of the North Eastern European borders. So we have different needs, different requirements, potentially even different technologies that need to be tested. So that's why we need to have this kind of satellite centers and through that we create a robust network where we can exchange knowledge, we can exchange lessons learned and we can all be better off. Of course this is done on a voluntary participation by Member States and of course there is also the potential for co-investment and joint exercises.
This is something also that we have been doing, table-top exercises, for example, but we can go also a step further than that. Next slide, please. I think that this is more or less explained. Our structure, the way we work at the JRC is to provide the knowledge we have through trainings, dedicated trainings. We did already one in Portugal in May 2026.
Research and testing. We have already some testing campaigns that have taken place in JRC Geel site. Our input to EU policy, the counter-drone action plan, and of course our handbooks and reports. These are, let's say, the pillars of our activities. Next slide.
And the proposed roadmap. So the short-term priority, as I said, is on unmanned aerial systems. Now we are in phase one.
We are signing an administrative agreement between DG HOME, actually signed already between DG HOME and JRC, in order to work on the, to finance the transition of the counter-UAS living lab to the center of excellence, establish a working group, and start and also developing the satellite site requirements. That's, let's say, the phase one in 2026. Now phase two, which will be versus Q1 of 2027, we will have some initial counter-UAS test campaigns and expand JRC Geel capabilities through additional UAS and an increased lab capacity. So it's about investing and bringing and really putting meat into this concept of the center of excellence. And the phase three, which is after Q1 of 2027, is to have the COE formally established, possibly with one, at least one satellite site activated.
And of course then we will have a plan about UAS operator training to be, let's say, fully operational. Next slide, please. I think that is over. I thought there was one more, but I think that we reached the end of the presentation. Thank you very much for your attention, and I don't know if you have any questions, happy to answer.
Speaker 3
Yeah, thank you, Mr. Giannopoulos. Let me turn now to the members and their questions.
Vasile Dîncu
We'll start with coordinators and political group representatives. Dear colleagues, please limit our intervention to maximum two minutes. We begin with Mr. Ruotolo, Sandro Ruotolo, from S&D.
Sandro Ruotolo
Grazie, Presidente. La diffusione dei droni ha stravolto il nostro concetto di sicurezza territoriale e, come stiamo osservando nei recenti scenari di conflitto, i droni stanno diventando sempre più autonomi grazie all'intelligenza artificiale. Quindi le chiedo: delegare decisioni letali a un'intelligenza artificiale pone un problema molto concreto. Un algoritmo non è sempre in grado di, come dire, di distinguere con certezza un bersaglio militare da uno civile.
Allo stesso tempo, per rispondere a minacce sempre più rapide, anche i futuri sistemi anti-drone dovranno inevitabilmente usare l'intelligenza artificiale per automatizzare e rendere più veloce la risposta. Mentre lavorate alla definizione dei nuovi standard di sicurezza, vorrei chiederle come il centro di ricerca intende prevenire il rischio di danni collaterali, soprattutto quando questi sistemi operano in aree urbane e densamente popolate. Esistono già protocolli chiari per garantire che anche in presenza di un elevato livello di automazione, il controllo umano sulle decisioni distruttive non venga mai meno? E possiamo avere la garanzia che una macchina non possa in nessun caso prendere in totale autonomia una decisione letale? Grazie.
Fabrice Leggeri
Thank you very much, Chair. I will speak in French. Merci beaucoup au directeur de l'équipe du JRC pour sa présentation. Je crois que l'actualité récente a montré en effet qu'il y a des menaces militaires par des drones.
On l'a vu même sur le territoire de la Roumanie puisque le président de votre pays a connu dans le port de Constanta et puis dans une autre ville voisine des arrivées de drones qui ont causé des dommages et qui auraient pu causer des victimes. On le voit aussi dans le Golfe et dans d'autres pays du Proche-Orient amis qui sont victimes des drones iraniens et on le voit bien sûr sur le théâtre d'opérations ukrainien avec l'emploi de drones notamment iraniens par les Russes agressant l'Ukraine. Donc il y a toute une dimension militaire et je comprends que évidemment votre mission est plutôt civile par rapport à des menaces hybrides. Mais est-ce que vous pouvez nous dire néanmoins avec toute la confidentialité qui est peut-être nécessaire si vous travaillez avec ou à la demande de certains États membres pour avoir une protection globale contre des drones qui peuvent être une menace militaire, qui peuvent être une menace simplement criminelle.
Je pense notamment en Pologne, il y a peu, je suis allé voir la frontière avec la Biélorussie, des petits drones sont utilisés pour du trafic de tabac et bon voilà, donc ça c'est de la petite, je dirais de la petite criminalité. Et puis il y a toutes les menaces hybrides pour perturber et qui peuvent aussi créer des dégâts, y compris perte de vies humaines. Donc je voulais savoir quelle est la coopération ou non ou l'interaction avec les recherches militaires de certains de nos États membres. Et puis je voulais savoir aussi si dans le domaine civil vous travaillez avec d'autres agences, je pense à l'agence Frontex dont la mission est d'aider les États membres à surveiller les frontières, et je pense aussi à l'Agence européenne de sécurité aérienne, EASA. Donc est-ce que vous pouvez nous dire quelque chose de votre travail avec ces agences civiles ?
Et enfin je voulais savoir quel était le budget que vous envisagiez puisque vous nous avez dit que c'était important par rapport au cadre financier pluriannuel, donc je voulais savoir un peu l'ordre de grandeur du budget dont vous auriez besoin. Merci beaucoup.
Helmut Brandstätter
Mulțumesc, mulțumesc, chair. Thank you very much. And yeah, thank you very much for all, for this report and all these details, very interesting. You didn't mention, if I'm right, where do they come from, who is guiding those drones ? You must have some intelligence about that, of course.
And the next question is, you talked about airfields and of course they're disturbing the air traffic, but as far as I hear in Germany, for instance, there should be a lot of drones over industrial sites of military production, etc. So there are some drones who just want to disturb, are there other drones who want to spy? And what's the background to that? I think that's really important. And then you said there would be you will have common practicing in Portugal.
Why in Portugal? As far as I understood it, the danger is more is more is way bigger in the in the East than in the West, but that's just a question about that. Yeah, that's it for now. Thanks very much.
Alexandra Geese
Thank you very much, chair, and thank you for for the interesting presentation. I have two questions. One is, do you work with member states at all? Because I mean, the information is very interesting, but you said yourself you obviously can't engage in mitigation measures. So how does that intersect and interlace with member states reflecting on mitigation measures?
Because that's really the problem we have, that we don't know what to do about it. And the second question is, since you in the JRC, you also work on disinformation, on the information space, do you work together to follow a little bit how these drone attacks intersect with disinformation campaigns? Because this is what we observe in Ukraine or also in Poland, for example, that the drone attacks are not actually used right now, especially in Western Europe, for military attacks. That might be different on the eastern border. But the main purpose is to sow distrust in national and local government, in the ability of democracies to provide security and safety for their citizens.
So it only makes sense if there is an accompanying disinformation campaign. Is that something that you're monitoring and where the different units of the JRC are working together? And if yes, what are the results? Thank you.
Tomáš Zdechovský
Thank you, Chairman. I collect all the questions from EPP members, also I have it here 120 questions, but I will not share all with you. I have to say at the beginning that Europe has good technologies for countering drones, but we still have one big problem.
It's much easier to develop the system than test them in Europe. Also many European companies tell us that they cannot carry out realistic tests because the rules are different in every member state. It's one very important thing. The procedures are slow and there are not enough testing sites. As a result, some companies test their system outside the EU, for example in Switzerland.
And it's not good for European innovation and also not good for European security. Also my first question from 125 is can the JRC tell us what the biggest legal barriers are? And do you support creating common EU testing site with one set of the rules? My second point is about the test this system. Counter-drone system must do more detect a drone.
They must also stop it safely. We know that Ukraine that jamming is not always enough.
Some drones are autonomous and fly without GPS. Europe must be able to test the whole system included hard kill solution in realistic condition. Is JRC ready to test the whole counter-drone system not only sensors and jammers? And finally Europe wants strategic autonomy. But many drones and counter-drone systems still use components from China.
This create security risk and it's also creates difference on suppliers outside Europe. We should know that we what is inside the system and we buy and where the critical parts come from. Europe should buy secure supply chain and support European producer. Does the JRC asset the risk of Chinese component and my last question do you think that EU needs stronger security rules and trusted supplier for drones used by our armed forces police and critical infrastructure.
Sorry for that that I was longer, but I promise that nobody from EPP will ask any question.
Vasile Dîncu
Thank you. Thank you, Tomas. Mr. Juan Fernando López Aguilar for the last question.
Juan Fernando López Aguilar
I adapt to some of the concerns that have already been expressed because it's only obvious that the aggression, the war of aggression by Putin's Russia against Ukraine has changed completely the state of mind of the European Union. For the first time ever, we are talking about European defense industry and joint procurement and joint action and an implementation to warmongering of the autonomic strategy doctrine. But the point is that it has also changed, obviously, the very nature of war itself.
It would have been unthinkable when I was a youngster that war would be like this, drones launched by countries waging war against each other, and that's what war is about now, drones, which is unmanned, unmanned war tools. The point is, first, in your analysis, is there any chance that drones capabilities are to be developed by individual member states, as it has been the case of Ukraine so far, or should it be a joint strategy altogether of the European Union in all when it comes to building up a drone capabilities? And second, I add up to the concern raised by drones going on their own, not subject to human oversight. So drones going out of control, what is the prospect of something like it happening? I mean, we saw when we were kids movies about nukes, nuclear bombs going rogue, out of control.
Is there a possibility of drones going rogue and out of control? What's going to be done about it? Thank you.
Georgios Giannopoulos
Very good. I think that it's not enough time to answer to all these complex questions and thanks that we didn't have all 120 questions from the EPP because I think we had to stay here the whole day. But no, definitely it's a topic that is extremely important, extremely at the edge of technology. And I will start from the first and last question actually, which are linked, and it is about artificial intelligence and autonomy. Our belief in the JRC when it comes to security is that there must always be the human in the loop.
This is our let's say ethical compass, if you want. So in the past when we were working on the conceptual model for hybrid threats and so on and also about detection of hybrid threats, what we always advocated for is that the human analyst and the human in the loop is the most important element. The technology is there to help to take a decision, but the decision should not be taken by the technology. That is what we have been advocating. So in terms of let's say the moral aspects of it, fully in line with what you mentioned, we have to have the human in the loop.
Now, of course, we don't define the environment always.
There might be actors who develop technologies that are fully autonomous. There we need of course to develop on our side systems that are in agreement with our ethical considerations, that they make use of artificial intelligence to speed up the decision making, but I strongly believe that the human in the loop will always be very very important and has to be there. It will be a challenge because we will need to take fast decisions and react very very quickly. And there are some nuances there when it comes to artificial intelligence. It is one thing to have artificial intelligence in a drone that finds its way without GPS for example or GNSS systems.
This is done with image recognition technologies and so forth. It is done today, but it doesn't take a decision what to to attack to. Actually this is has been already pre-programmed by a human actor. So this is our position for the time being.
Of course, we cannot guarantee that some actors will develop this kind of capabilities, but then we need to be to act very quickly from our side, both at the legislative and also at technology level, to be able to counter that in line with our ethical considerations. Now, going to the question and also complementing the last question, which is about if there is any possibility that drones go by their own. Yes, now they can go by their own, but as I said, the decision what to do has been already taken by a human in the loop. Now, when it comes to civil or military aspects of drones, actually in the JRC, we don't work fully on defense. We work on what we call dual-use technologies.
And practically everything, almost everything today is dual-use technology. It can be used both in the civil and in the military domain. We collaborate with the military domain. For example, we just signed a memorandum of understanding with the European Defense Agency so that we can exchange views and let's say data, if possible, with them so that we can facilitate our research, but also we can provide part of our research results to the member states through EDA. We work a lot with civil organizations.
Actually, this is our bread and butter, if you want. We have a working arrangement with Frontex that we have signed and we work very closely with Frontex.
Actually we have done in a few weeks ago in our JRC Geel site a demonstration event with Frontex on detection technologies for border security. Doesn't go is not directly linked to to drones but it is connected to drones of course. We work with EASA for the aviation safety because drones are also a major concern for aviation safety. And your very important question about the budget for the center of excellence. This is something that we don't have a figure if you want for the time being for the next MFF.
This is something that we are now putting in our let's say proposals. There are of course limits to to what we can do in the sense that we are we have our staff reference table and we cannot expand infinitely if you want but certainly more than what we have now will be needed so that we can offer a sustainable research if you want and solutions to our stakeholders. Your question about there was a question about intelligence and what intelligence says about drones who's behind that.
Actually, this is an area that we don't work, so the threat analysis, the threat assessment is not done by us. We are in, let's say, we are entering into this whole cycle if you want from threat analysis to mitigation of drones after the threat analysis. So this is done by the member states or by INTCEN, for example, they have information, they define the threat somehow, the threat environment, but then we, this is an input for us and we are not working on this particular point. And there was also the question why we have this training activity with Portugal for the time being. That is simply because we got the request from this member state.
Of course, we are open to all member states to do similar thing. We need, of course, if we get request from all 27 member states, we will have to prioritize because of resources and do it in the course of a certain time period, but for the time being, this was the first request we had and that's why we responded to that. When, of course, we work with member states, that was another question. We work very closely with member states, in particular with research organizations. We have signed several agreements recently which include also drones.
And very, what you mentioned about disinformation and the link with drones, that's a very interesting point. This is really at the core of what a hybrid threat is. There is an incident somewhere, a cyber, drone, terrorist attack, whatever, and then it is exacerbated through the media so that the actors can demonstrate that the government or the authorities cannot take care of the situation, they cannot provide security to the citizens, and there is a mistrust in the whole system. And this is the overall objective. Fully share your opinion on that.
Actually, what we have done at the JRC, and this is a very interesting discussion actually, is that security in the past was mainly a technical topic. When we started working on drones, we were doing some simulation to see what would happen in case there is an explosion by a drone in a specific area. Very technical topic. But we have expanded the scope of this and now it becomes something that we, it's much more comprehensive, it's the whole-of-society and whole-of-government approach. And for this reason, we have developed the portfolios at the JRC, the portfolio approach.
So we have 25 portfolios that deal different with different topics, under which we have several projects, exactly so that we can see the connections between the different topics.
And of course, in the security domain, we are working with our colleagues in the disinformation environment. We have tools like EMM, for example, which can monitor the narratives that exist out there. And we try to connect the dots and try to see if this is really part of a hybrid campaign. It's ongoing work that we we do. Of course, there was there was a question on the testing.
Yes, we need to test at a certain point hard kill solutions. We cannot do this at the JRC for the time being because the legal framework is such that does not allow that. Indeed, there are not enough testing sites and this is one of the shortcomings that we try to tackle with this center of excellence to offer the capability to do that. Also because our commissioner's mandate is startups and scaleups, Commissioner Zaharieva. And what we try to do now is to open our research infrastructure and testing facilities to startups and scaleups so that they can come and test their technologies.
Because exactly they don't have the capability to build such a comprehensive testing environment for their technology. So we are opening that. The we tackle drones as a technology stack, which means that we need to see all the supply chain of of drones. I don't think that we will ever be completely independent from third countries when it comes to supply.
It takes years and time exactly to build this capability. We are aware of what is the the challenge there. We are working, I think with different instruments in the European Commission, in the European Union to find a solution into that and to build let's say the capacity to have our own components inside Europe, but it will take time. It will not happen from one day to the other. And there was also one question about the drones capabilities, whether these are developed by single member states or by many member states.
That's a very interesting question and I think that goes back to the threat analysis. Different member states have different threat landscape and different threat understanding and the threat environment and the threat understanding defines the kind of capabilities that each member state would like to have. Of course there can be a coalition of member states who understand the threat environment in a similar way and then they need to develop similar capabilities. But definitely if we have more than one member state developing a technology, and this is actually the idea of many of the of the schemes that we have now, financing schemes, I think that we can achieve much more. This is I think clear.
And I think I'll leave it here. I think I went through all the questions.
Thank you very much.
Speaker 13
Thank you, Mr. Giannopoulos, for your explanation. Thank you. Okay. We continue in Romanian.
Vasile Dîncu
Please prepare for translation. Următorul punct al ordinii de zi de astăzi este prezentarea raportului JRC Realitatea fracturată: cum alimentează algoritmii polarizarea și afectează democrația. Vă amintiți că în aprilie am primit JRC pentru a prezenta raportul privind provocările viitoare ale democrației, studiul care a analizat presiunile asupra democrațiilor noastre la nivelul cetățenilor, instituțiilor și mediului informațional. A fost un studiu apreciat de membrii noștri. De aceea astăzi revenim la această a treia dimensiune, spațiul informațional, cu o analiză mai detaliată.
JRC a publicat un studiu dedicat acestei teme numit Realitatea fracturată: cum poate democrația să câștige în lupta globală asupra spațiului informațional. Suntem bucuroși să-l avem astăzi pe unul dintre autorii acestui studiu, domnul Sharp-Billing. Vă mulțumim pentru prezență. Raportul susține că problema nu constă doar în afirmațiile false pe care le putem verifica și demonta.
Adevărata provocare este modelul economiei atenției care recompensează conținutul negativ, emoțional și conflictual, ceea ce autorii numesc complex industrial al fanteziei. Scopul acestui model nu este neapărat să ne facă să credem o anumită minciună, ci să ne epuizeze și să distrugă terenul comun de care are nevoie orice democrație, aș spune mai departe, certitudinile pe care le avem. Raportul JRC arată că educația media și verificarea faptelor funcționează, dar sunt eclipsate de modelul de afaceri de bază. De asemenea, normele noastre au un impact limitat, deoarece cele mai mari platforme sunt deținute și controlate din afara Uniunii Europene. Raportul conchide cu concluziile noastre privind necesitatea de a aplica reguli existente, în special în ceea ce privește regulamentul privind serviciile digitale, DSA-ul și regulamentul privind piețele digitale, DMA-ul.
Este esențial să eliminăm blocajele la accesul de date care îi forțează pe cercetători și chiar pe Parlament să lucreze din lipsă de informații. Trebuie de asemenea să protejăm integritatea alegerilor și să susținem mass-media independentă și societatea civilă ca element de pregătire. Există un punct pe care aș dori în mod deosebit ca oaspetele nostru să-l detalieze. Comisia a propus și a creat un Centru European pentru Reziliență Democratică.
Cred că în februarie a fost lansat acest centru. Raportul nostru salută inițiativa, dar exprimă temerea că acesta ar putea deveni doar un alt centru fără putere operațională. În mod surprinzător, JRC-ul solicită crearea a ceea ce numește un CERT pentru date și democrație, o capacitate europeană comună de a studia platformele la scară industrială și a testa intervenții în timp real. Aș dori să aflu dacă considerați că o astfel de idee ar putea fi preluată în viitor de Centrul pentru Reziliență Democratică. Domnule Scharfbillig, aveți cuvântul pentru aproximativ 15 minute.
Stephan Scharfbillig
Mr. Chair, honorable members, colleagues, many thanks for inviting us to talk about our research on fractured reality.
I'm leading a research program in the JRC on the Enlightenment 2.0 where we have for many years now looked into polarization of society and solutions to this, but also what does trustworthy public communication mean in this context, how should we talk to each other, how should we debate, how should we move democracy forward. I want to start by saying that with this report we work with an interdisciplinary group of world-renowned experts in information integrity, platform regulation, data science, and psychology to understand what is really going on under the hood, what does it do to our society. And therefore we came up with this title of fractured reality, where I want to point out there are other terminologies that are out there, for example like fragmentation, but we really felt fractured is the right term, right title here because it shows more this intent from the outside in algorithms fracturing our reality and also showing this conflict that it creates within our society. It's not just fragmentation that we are peacefully in different bubbles, but it's actually an issue of conflict that is created. Next slide, please.
I want to start by saying why this matters to us.
The chair already mentioned the report by our colleague Mara Almeida who presented this in a few weeks ago, that our understanding of democracy is not just that we vote every now and then for the colleagues here, but rather that it needs to be filled with life, debate, pluralistic debates that are had in the parliaments, in society, amongst citizens, that needs to be based on some shared sense of reality, on information that is robust and that we can then determine our future together. But we see that this is usually ensured by information integrity, that people have agency over this, finding information that is truthful and consistent and that we can come to joint understanding. The information space has been changing dramatically over the last decades specifically, and that is the baseline of or the source of why we are thinking this research is much important. And we see the symptoms in society already. We see increasing, especially affective polarization, polarization between people disliking each other, it's not even on the content, it's about disliking each other.
And we see one interesting study by Edelman Trust Barometer showing that we have also seen a dramatic collapse of people saying that they're no longer reading information from other politically leaning sources, such that we can no longer even understand what the other side knows and sees. Next slide please. So we looked into the evidence and found three main ingredients to this that we should need to consider. The first is information overload. Information overload means that, and there's one number that we can put to this, a 20% increase of information online available every year, year on year for the last five years.
Certainly AI is not going to decrease this information availability. And normally we'd say, okay, more information is good, right? We want people to have access to this information. However, psychology shows us that when people feel information overload, they're focusing actually on a smaller set of information they already care about, a smaller set of topics they care about. They also pay more attention to emotionally charged content that refers to outgroups, other groups in a negative way.
And finally, they're losing the ability to make meaningful connections between causes and results because they're so overloaded by everything that's going on that, yeah, this meaning and this sense of understanding of what's really going on is decreasing. The second is echo chambers and even echo platforms. I think I don't need to explain to the colleagues here in this room what an echo chamber is. The research on this has been somewhat contested earlier, but the evidence now is relatively robust to say, yes, they exist. It means that people receive information on platforms, primarily through algorithms, in a personalized manner.
There's an estimate about one third is kind of pre-selection based on the algorithms and two thirds by the engagement of the users themselves.
And this is not new, but it's just turbocharged by algorithms and made much more relevant, much more important. We even see this move towards echo platforms, meaning people go to different platforms altogether and cannot even interact with each other anymore. And this is a problem specifically because of the third and central factor here, that it also distorts our reality, our sense of what is true and what the other people see. These echo chambers are specifically there and running with this kind of emphasis on showing content from other sides in a toxic, negative, and emotional way. So meaning people see content from other politically leaning groups, but they always see it through the lens of negativity.
This is what the others think, look at how ridiculous this is. And these three things ingredients together lead us to the fantasy industrial complex, next slide please, where we see now over the past 20 years we have a shift from there was always debate about truth and evidence and what is real, what is not real, to the I mean where evidence was still discussed where everyone tried to figure out what is real, what is not, and once people were presented with overwhelming evidence of some facts, usually there was this agreement, okay, then we accept this.
And now we have moved towards this era of constructivism where we can perfectly hear and see these these terminologies of alternative facts, of yes, you have your truth, I have my truth, and where where reality is then determined by as we call it this fantasy industrial complex, which is kind of a loose group of where influencers, media, alternative media, but also politicians can interact with each other and create realities that are no longer actually grounded in any reality. And so together this leads us into not only that we have a fracturing of reality where people see different realities, we have different camps that hate each other and and dislike each other, we have experienced this in many ways, but also a kind of divorce from reality altogether that can be seen on on multiple areas. Next slide please. What is the fundamental driver of this is is the important question that we asked. And there's some good evidence to suggest that it's really the business model fundamentally driving this.
The attention economy is is key to this. There's one study that looks at at this kind of echo chamber and polarization effect across different platforms that have different business models. And showing that those platforms that are run on a personalized feed that needs to need engagement because they want to sell advertising, that leads to this extreme echo chamber effect. Whereas other platforms that run for example on subscription models or have this community feature such as Reddit, that they are not having these effects. And so it's really an engineering question.
What is the algorithm trying to do and what is the business model behind it how they make money? Now there's other studies showing that yes there's indeed this conflict and trade-off for businesses then if you run on engagement models between toxicity and and revenue such that if you reduce toxicity people engage less with the platform and less with advertising. So it's a clear business conflict here for our democracy. Why is there no solution organically emerging? Well we see that people are locked in through network effects.
If many people are on a platform there's a lot of content then just moving to another platform is not that good idea for many people. There's also this indication of addictive design meaning or habit formation depending on how you want to use the definition where one estimate says that about one third of the use for example of Facebook in one study is can be notched down to just a habit, just going there regularly if you call it addiction but without any any personal use to these people. Usually users are aware of this.
This is not necessarily news, I mean, this has been everywhere. But they also say in studies that their autonomy over the choice is extremely restricted. If they want to get out of this, the the platform design and the interface always leads them back to this personalized feed. So it's really hard for them to change. And one indication that I find most telling is in one very clever study, people were willing to pay money to turn off social media for their friends and themselves.
So if you ask them, do you want this to be turned off for you individually? They say no, I need I mean, I need to be there, other people are on there, I need to be informed what's going on. But if they have the ability to say yes, if it can be turned off for my friends and me, then I'm willing to pay money for that. And that's I think the most telling of where we are with social media right now. Next slide please.
How can we arrive at better democratic business models? I I don't have too much time to go into all these these issues, but we really think that the evidence points to the fact that we also need some degree of digital sovereignty and autonomy over our information space. Obviously we have sovereignty, we you make laws together in the EU on this, but the question is are they are they really followed.
And I want to focus on one of these threats here as a very specific point, namely the indirect interference in our democratic system. For example, there's one study on the war on news by Facebook, where they showed that over years, Facebook has reduced the engagement with news on that platform. And the study looks into this and shows that it's not that this is a general reduction of supply of news, there's no lack of overall engagement. It's just a reprioritization of the algorithm on news and based on their business model of what they find the most relevant. And this is foreign control over the quality of information that people see in our democracies across the globe.
And we think that we need to have options over this to change this, and it will not organically emerge. So next slide please. To think about this in a more structured way, we think the most important thing is to recognize these levels of the issue. We see and we are faced with ourselves as users of social media, various platforms, this daily interaction with toxic content, debate, fake news, disinformation, misinformation, and all of this. But that's on the surface level of how we interact.
I told you about the key ingredients, you know, information overload, these kind of echo chamber effects, distorted realities, the fantasy industrial complex. And we can do something about this. But the the more important point is to understand this is coming naturally out of the dominant business model that's currently out there, namely this engagement model, and that it's a technological choice, that there are alternatives and we see that if this is not the business model, that yeah, it works differently. And then the top level of the geopolitical dimension to this. So how we structure our solutions to this is to say, well, first of all, we need to we think it's it's prudent to invest more into digital sovereignty on a digital, physical and online space, infrastructure for this, to also support decentralized online platform alternatives.
We may make some references to the atmosphere, the Fediverse, because these are easier to be split into the different individual parts of it. It's not only the moderation system, it's not only the content surfacing, the interaction, the social graph. All of these can be separated into different parts of it and thereby make it much easier for us to to regulate and to understand how it's working.
And we should, as institutions at the EU level, but also at the member state level, go ahead and be the role model for this. We should do our own communication on these platforms, or could do our own communication on these platforms, do the procurement in line with this, with our demands here for democracy, and therefore enable this. Then on the business model level, we also have several recommendations there of how to achieve better, I mean, change the business model. There are several Nobel Prize laureates in economics who suggested, for example, taxes on advertising that would shift the incentives towards, for example, subscription models and others. But even if we don't want to do that, we can also think about how we can get algorithms that work for democracy by, for example, encouraging interoperability and in situ data rights, which would allow a market for middleware to exist and to emerge that then would offer the choice to users rather than us telling them what's the right solution there.
And then finally, there's also some experimental views on this, for example, speech markets where you can put your money or reputation on the truth content and then citizen panels can determine the truth such that they show, at least in experiments, that it works really to increase engagement of the center, of the middle of society who are otherwise silenced at the moment by these platforms. And then finally, if we have done this, then on the cognitive level, on the direct interaction when we talk about disinformation, FIMI, all of these, there's still these questions about fact-checking, pre-bunking, de-bunking, all of that that we think is useful and it works. We see, for example, media literacy in a meta-study having done various different interventions, media literacy is the most effective in these things. But yeah, these are only really effective if we address these other issues at the top. And with this, I would like to thank you very much and I'm open to your questions.
Vasile Dîncu
Thank you, thank you for this scientific and very interesting report, very important for us. Please, dear colleagues, please limit our intervention to one minute and to question. We begin with Mr. Sandro Ruotolo from Italy.
Sandro Ruotolo
Grazie, Presidente. Grazie, dottore, per il suo prezioso intervento.
Mi ha molto confortato perché sono due anni che noi ci battiamo per la sovranità digitale, per l'architettura, per gli algoritmi, per la trasparenza. Io pensavo al peso delle piattaforme sociali e se si ricorda alle elezioni della Germania, il signor Musk, forte dei suoi trecento milioni di follower, fece la, come dire, l'intervista alla leader dell'estrema destra dell'Alleanza for Deutschland. E oggi, e oggi il signor Musk, come dire, alla vigilia delle elezioni francesi, come dire, fa l'endorsement per la signora Le Pen. Quindi pensate che cosa possono fare queste piattaforme. Noi abbiamo fatto degli studi sulle elezioni svolte negli ultimi anni in Europa e mostriamo, mostrando che la viralità da sola non basta a spiegare la sovraesposizione ottenuta da alcune forze politiche e da alcuni contenuti più radicali.
La distribuzione della visibilità politica dipende da sistemi di raccomandazione opachi e controllati dalle piattaforme che possono autonomamente decidere senza alcun controllo democratico quali attori politici rendere più visibili durante una campagna elettorale. Chiudo dicendo questo.
Avete il codice dell'algoritmo o i dati di accesso per verificare subito se le piattaforme applicano a tutte le forze politiche gli stessi criteri di visibilità? La nostra sensazione è no, che non esiste questa trasparenza. Quindi chiediamo da questo punto di vista, lei lo sa che la Commissione europea ha aperto un'indagine, noi chiediamo una commissione d'inchiesta parlamentare perché il Parlamento deve esserne a conoscenza. Abbiamo bisogno della par condicio digitale.
Virginie Joron.
Virginie Joron
Oui, merci, j'avais juste un quiproquo sur le temps de parole, vous savez ici c'est très cher, une minute, deux minutes, ça compte beaucoup dans les discours. Je vous remercie monsieur pour votre présentation. Donc on voit aujourd'hui effectivement, mon collègue en parle un peu, monsieur Thierry Breton vient de tweeter : "La saison des ingérences commence". Elon Musk soutient donc Marine Le Pen, c'est son droit, charge aux autorités de veiller à ce que l'algorithme de X ne favorise aucun candidat.
Musk effectivement vient de déclarer sur Twitter que Marine Le Pen est le dernier espoir de la France et il a bien raison. J'ai lu votre rapport réalité fracturée. Comment les algorithmes alimentent la polarisation et affectent la démocratie. En effet la commission est une fracture de la réalité. Vous décrivez un complexe industriel de la fantaisie avec ses algorithmes de l'économie de l'attention qui enfermeraient les personnes dans l'écho de leurs préjugés avec une désinformation qui ne vise plus seulement à faire croire des mensonges mais à générer des réflexes antidémocratiques.
Intéressant. Un peu comme ce rapport lunaire du Sénat français du centre mou et l'RPS qui invente le concept de lutte contre les ingérences intérieures. Vous écrivez qu'il y a 10 à 30 % de désinformation sur les publications des sujets litigieux climat santé. Est-ce que vous avez aussi compté la désinformation de Bruxelles et de Pfizer car vous ne parlez pas des 95 % d'efficacité. Vous nous dites que les populistes de droite désinforment.
Votre source une étude bidon, oui je le dis, l'étude 157 qui utilise Wikipédia et Media Bias Fact Check, source américaine qui note les médias mais qui ne connaît pas TF1 premier média français. J'ai deux questions. Expliquez-nous ce que vous entendez par contenu polluant. Est-ce que l'esprit critique est remis en question ? Deuxième question.
Si vous tenez compte des statistiques sur l'utilisation des réseaux sociaux en France ce n'est que 8 millions d'utilisateurs sur X.
Nous sommes tous juristes ici, nous écrivons des lois tout le temps, alors j'ai envie de vous demander comment vous matérialisez les effets des algorithmes sur la finalité du bulletin de vote. Je vais vous donner un exemple. Pendant la période COVID, toute la journée je regardais BFMTV, toute la journée on me demandait d'aller me faire vacciner et je ne me suis pas fait vacciner. Merci.
Chris MacManus
Merci. Thank you, Chairperson. Go raibh maith agat. Thank you very much, Mr. Scharf-Bilig. Excellent work.
Very insightful report, very worthwhile, but also deeply concerning. One point that I strongly took from it is that it's not just about addressing misinformation and disinformation. The report quite rightly pointed out that addressing misinformation and disinformation is only a symptom of the problem. And the deeper problem is that people are increasingly living in a separate perceived reality with less shared media and less shared cultural and factual common ground. Social media algorithms, you point out, they intensify this by showing different groups polarized and emotionally charged content designed to maximize engagement based on a business model, as you've eloquently outlined.
So as regulators, Mr. Scharf-Billing, how can we address that underlying system rather than only correcting the individual falsehoods? And in fairness, I do understand from your last slide that you had many suggestions tiered, but how can we preserve freedom of choice and political pluralism whilst ensuring that citizens are exposed to a fair range of credible viewpoints? And finally, is there anything that we can learn from traditional broadcast media, a world that I worked in for many years? Traditional media in most liberal democracies, it exposes people to a wide range of topics and interests and perspectives. Are there any models or methods that can be adopted from the traditional media for the new social media?
Thank you, Mr. Scharf-Billing. Thank you, Chair.
Alexandra Geese
Yes, thank you very much, Mr. Scharf-Billing. I also really commend your study. And what I really appreciate is that you don't stay at the surface. We discuss a lot disinformation, but as my colleague said, it's more a symptom than a cause, actually. And you very correctly said the cause is engineering.
This is engineered. It's not that people, you know, are so much more interested in that kind of news or it's really an engineered system. And you said what the three, the three key factors are. It's engagement.
It's engagement, it's engagement-based ranking. This is, I think, the worst, combined with the profiling, so the platforms exactly know what kind of content send out to us to keep us engaged based on our profiles. And the third one is addictive design, which doesn't allow us really to go out. And then additionally, you mentioned this sort of lock-in effect, that a lot of people say, "I would leave if my friends left as well." And this is even worse in politics because this is also very true for us.
So if we don't abide by the engagement-based ranking criteria, which means we have to have out-group hostility, because this is what works best, and you mentioned that, thank you for that, because a lot of people don't see the connection, we are not visible. And that means our parties don't get votes and we don't have power.
Where I think the study falls a little short is in terms of solutions, because I share absolutely everything you said, but it doesn't go far enough because it doesn't really address the lock-in effect and the addiction effect. And I think under the DSA, which you did not mention unless I missed that, under Article 34, it really, it very clearly says if there is systemic risk for public discourse and elections, for freedom of information, for freedom of expression, and this is what we have, because everybody who does not use out-group hostility, for example, has a lot less chance to get, to have visibility under engagement. So what we really need to address is to use DSA in order to say engagement-based ranking is a systemic risk. We have to give people back their choice. People have to be able to choose, and engagement-based ranking is not people's choice.
So I think this is really the most important remedy. Then reduce the profiling, we also have an instrument in the DSA to at least reduce it, and we need additional instrument to stop addictive design, and the Commission is already at least investigating on this. So I think this is the most important measures, and then all the others will work. Would you recommend that? Thank you.
Tomáš Zdechovský
Yes, thank you very much for your study and for your overview about the situation how the algorithm is working. I have to say that we are facing a lot of disinformation through the Russia or Chinese bots. And when I was in the military department of cybersecurity, we saw there increasing of the fake accounts what are really spreading the European democracy and how they were active in very important days for democracy. It's not only the elections, but it's also all the referendum or also very important debates about the future of the state or future of European Union. My question is simple: how really stop them?
What we can do? Also is any recommendation, clear recommendation what we can do for the future? Also this is my first question. And second questions: really, we look systematically how some extremists from the right and from the left are supported by algorithm of of some some platforms. How to stop them?
Thank you very much.
Csaba Dömötör
Let me speak in Hungarian.
Reagálni szeretnék erre, tehát reagálni szeretnék az Európai Néppárti felszólalóra az imént. Tehát ez egy szokásos jelenség, hogy amikor önök veszítenek a támogatottságukból jó pár országban, akkor hirtelen elkezdik az egészet ráfogni az algoritmusokra. És nem állnak meg itt, mert különböző tényellenőrző rendszereket erősítenek meg az Európai Bizottsággal összhangban, hogy lejjebb tekerjék azokat a hangokat, amelyek önöknek nem tetszenek, és amelyekre egyébként van igény, mert a választásokon kiderül, hogy állampolgárok tízmilliói támogatják ezeket a pártokat. Ez nem demokratikus. Tehát arra kérem önt, hogy ne annak a lehetőségeit keresse, hogy megpróbálja elhallgattatni azok hangját, akik nem úgy gondolkodnak, mint az itteni nagykoalíció.
Köszönöm.
Helmut Brandstätter
Yeah, thank you very much. I basically have just one question, and Tomas talked about fake accounts. We do know, and when I follow especially on X and on Facebook, I see those fake accounts. And you see that very often they just repeat what is already said. And we know that this troll fabric in Russia, we had here Philip Obaji from from Nigeria, and he told us how the Wagner Group in Russia is also in Africa is also organizing those.
So what would be the technical possibility to just shut them out and what is the juridical possibility to shut them out? Because I don't think that anybody here, included the patriots, do want that some people sitting in Africa get money just to repeat something which someone said, because that has nothing to do with democracy. That's the contrary of democracy. One man, one vote, or one woman, one vote, one woman and one man, one opinion, that's fine. But one bot, one vote, that's strange.
Stephan Scharfbillig
Perhaps I may respond to the questions and I think that may solve some of the issues. Many thanks for your interest in the studies and I don't know how long I have to speak but I'll try to address as many as I can. I think one of the core discussions here is centering around how, I mean, is disinformation and misinformation a core issue or what is the system underneath the problem? And how can we address the system underneath? I think in our report we agree very much that it's about the structure of the platforms that we need to address and the solutions that are there and we discuss quite in detail.
What we say there is really the importance of the changing the business model fundamentally. And even if we have sovereignty over the IT infrastructure, the models, if we had our own social media network in Europe, if it's run on the same model, it will result in the same thing.
And this is not about necessarily prioritizing certain political groups, it's about the type of content that is prioritized. It's the prioritization of disagreements, of the complete, yeah, doesn't matter the information quality basically to the algorithm as long as it creates and locks us in. And I think for democracies this is extremely dangerous to have as a model. What we propose there, the DSA, I haven't mentioned this in my presentation, we mention it in the report. Certainly many of the studies that we were able to review did work with data that came in prior to the implementation or enforcement of the DSA, so we cannot make at this point a judgment whether this is enough or not.
However, we think we shouldn't stop at thinking can we turn back or go back to an older model. We also make several suggestions to create alternative public spaces, to have citizen engagement in a way that is at scale at the European level, at national and member state level, where we also have a report just published, colleagues in our team on the impact of citizen engagement.
I think that would also be quite interesting and we reference in our report to this group, where civic tech would allow us to have meaningful debates where we could agree on common ground and then find a solution forward because it's an engineering space again. As a reminder, there are other options of technology that work to find common ground rather than to disagree, but this is not how these platforms are designed. There were, I think, questions about data and also from the chair previously to think about our suggestions on the CERN to have a large-scale infrastructure to analyze data. We definitely see from scientists, and there was a recent report published at the same time as our report by colleagues who looked across multiple countries in the EU, outside of the EU, whether platforms are following the obligations of providing data access and what data was provided and what length and speed, and usually find it lacking. And this is an extremely important point.
We know that the European Commission has put out a call for an infrastructure to facilitate this research in the future much more. I think this closes after the summer this year, and certainly we want to be a part of this and help make this work so we know much better what's going on.
At the moment, a lot of these insights that we report or have in the report, we also mention this explicitly, are usually not with in collaboration with the platforms, but in spite of what we have access to. There was another question of how to stop foreign interference accounts, bots, whether they are bots or not, or whether they are real people who do this. We do know, and I think we reference this in the report, that there's an incentive to create polarizing content, whether these users are in the EU or not is not relevant to this. And therefore, again, it comes back to: is this the type of content that we want to be prioritized in our online information spaces? And our solutions on the platforms, I think, speak to that, rather than having to verify this independently.
I think there's also my head of unit who wants to respond to the bigger question on the enforcement part. But I think ultimately we have the technical capabilities, but if we don't invest in an alternative way of structuring the information space, then we will not have the solutions. And pre-bunking, de-bunking, media literacy, putting the emphasis on citizens to do all of this while they're exposed to this fracturing of reality is not the right solution. I think that's what we make clear in the report.
Speaker 22
Thank you. David Mair from the JRC, just to pick up on some of the underlying issues.
It's important not to see this issue only about information environment technology. It's part of the wider whole of democratic resilience. So, as my colleague referred to, we talk also about it's important to have face-to-face conversations between people. So the problems that come from the online space for our democracy will not only be solved online. We need to get back also to making sure that the public spaces, the face-to-face conversations in our societies happen so that people can find each other.
And also the question of, I heard an underlying assumption that there is some sort of technological fix to what is going on. I think it's important to understand that there is the supply of information, but there is also the demand of information. And probably the surest way to ensuring these sorts of things are less problematic in our societies and we can continue to have intense disagreements, intense debate. Disagreement and debate is what democracy is all about.
One of the ways is ensuring that citizens have greater trust in democratic institutions, they have greater trust in government, they have greater trust in news sources. The more there is trust, and trust needs to be earned, it can't be granted, then hostile disinformation from whoever it is, from whatever side of the political spectrum, will cease to matter as much because people won't be influenced by it. So the underlying systemic issue we need to focus on here is building democratic institutions that earn and win back the trust of citizens.
Speaker 23
Thank you. Următorul punct al agendei noastre este prezentarea documentului de informare Impactul rezumatelor și a prezentărilor generale generate de AI asupra veniturilor și asupra libertății presei.
Vasile Dîncu
Suntem în întârziere și n-o să mai fac o prezentare, așa încât îl rog direct pe profesorul Nicola Lucchi pentru o prezentare de șapte minute. Please.
Nicola LUCCHI
Thank you. Thank you, Chair, and thank you to all the members of the committee. It's a privilege to be here today and grateful for the chance to present this brief to you in person. So, I would like to start with something very ordinary, something that we all do many times a day. We open Google and we search for news.
And what I want to show you is simple: that every day, that has changed, and the change without anyone telling us. More and more, the answer that we are looking for appear right there on Google home page. And the click that used to take us to the newspaper, to the publisher, or to the people who actually did the work, that click more and more often simply doesn't happen. In my written brief, I was much more technical on how this feature work.
Now today, I will not go into all this technicality. I would rather stay on the point that matter most to your committee. And specifically, what does this mean for journalism? What does it mean for media pluralism? And what does it mean at the end for our democracy?
So, let me start with the first slide, next slide. So, this is on the screen is what we are actually talking about. Is a real Google AI overview from a search that I ran a few days ago when I prepared this presentation. And look at how the page is organized. Right on the top, an AI summary answer the question for you directly.
Further down, the sources sit in a small side panel. And the first link to a real publisher, well, you have to scroll a long way down before you reach the first link. But here is the part that I think is most interesting.
You do not have to take my word for explaining the meaning of this feature. When you ask to the AI overview what is exactly it able to describe itself, it says exactly that it puts AI summary at the top of the search result and then, in its own word, that it saves you from clicking through the other website. So, that one sentence, I think is the core of my brief. The feature is specifically created, designed to keep user on Google page. This is exactly the purpose.
Next slide. Let me show you how the value move, because it's much easier to see visually than to explain. We can organize this in four simple steps. It starts with a publisher who invests real money and real effort in original reporting, in editing and in getting facts right. Then that publisher permit Google to index the work because that is how you stay visible in search.
This is the old bargain and it worked for everyone. Step three, if Google finds the information useful, it provides the answer directly and shows this answer on the top of the page. And so, step four, the reader gets what they need instantly without clicking on the link. But on the other side, the publisher never gets the visit. And this is exactly the old story in a single line.
And so far, so good.
The value moves, the attention moves, and the advertising moves. So, quietly from the publisher website to the search page itself. Next slide, please. Now let me turn to the evidence. And here I want to be honest with you about where we stand.
When I wrote the brief, the evidence was mostly observational. It came from analytic studies and from watching how users actually behave on the internet. And at that point, there was really only one proper experiment, so I chose my words very carefully in writing the brief. But now the picture is changing.
We are moving from correlation to proof of cause and effect. And this is happening quite fast. To give you an idea of the scale, Pew Research found that AI overview pushed publisher clicks down from 50% to 8%, that is almost half. And now new experiments are proving actual cause and effect. In particular, a recent field experiment showed that when AI overview appears, clicks to publishers drop by almost 40%.
Other studies are coming and they are already pointing to the same conclusion. Looking ahead, media leaders themselves expect referrals from search to fall down by around 40% in the coming years. So the picture is becoming consistent. Different methodology, different researchers, but all they are pointing out the same way, that is AI summary reduces the traffic that reaches publishers.
Just one final warning, this feature of course arrived in Europe later than in the United States, so today these numbers are only the tip of the iceberg. Next slide, please. Now, why we should, why your committee need to care about all of this? Because this goes well outside disputes over copyright or revenue. At its core, this is about media pluralism.
And media pluralism is what keeps a democracy strong. Specifically, I see in my brief two risks here. The first risk is concentration. Nearly 40% of the sources cited in AI overview are already in the top 10 search results. So instead of widening the field, this summary simply reinforce the player who are already dominant in the market.
The second risk is more delicate, call it flattening of perspective. On almost any topic, there are many legitimate way to tell a story. An AI summary compress them all together in one single answer.
And the publishers who feel the most side effects are the smaller ones. So, for example, regional voices, minority language outlets, and this is exactly where Europe pluralism stays. Next slide. And this connects directly with your own work. The text you adopted last month already recognized these type of risks.
Traffic loss, gatekeeper platforms, the pressure that artificial intelligence puts on copyright and fair remuneration. So these findings no longer come from outside. It's already in your own text, and that changes the conversation. Yeah, yeah, I'm just at the end. Next slide.
So how to respond? Let's be candid, antitrust cases are helpful, but they are too slow to solve this alone. I see three gaps. First, the output trap. Today, publishers face a false choice.
They are not able not to stay out of these systems. Second, remuneration. AI summaries use publisher work, but nobody pays for it.
And third, the gap is media freedom. The Media Freedom Act protect against content removal, but it say nothing about silent traffic diversion. I can conclude here. Thank you so much.
Vasile Dîncu
Thank you, thank you, Professor Lucchi. Please wait for our question. We begin with Sandro Ruotolo.
Sandro Ruotolo
Eccomi, rapidissimo perché siamo coi tempi... Beh, concordo con il lavoro che ha fatto, professore. Con l'intelligenza artificiale overview, il pluralismo viene sostituito da una sintesi prodotta da un soggetto privato, in questo caso Google. Oggi nessuno conosce i criteri con cui questo strumento di risposta semplificata seleziona, combina e gerarchizza le fonti. Se i cittadini ottengono sempre più spesso una risposta completa da Google, apparentemente corretta, senza passare per le fonti originali, secondo lei è più grande il danno economico ai giornali, per esempio, o il rischio associato alla qualità e al pluralismo dell'informazione?
Mi taccio per correttezza.
Speaker 26
Go raibh maith agat, a Chathaoirligh. Thank you, Chairperson. Thanks very much, Professor Nicola.
Chris MacManus
Yes, local media is facing more threats every day, as you've just outlined, and Google AI summaries, they're appearing to replace its traditional search model. If someone hears about a local incident and Googles it, it's no longer the local newspaper they see first, instead, obviously, as you've outlined, it's going to be the AI-generated summary. And there've been numerous cases of news organizations criticizing Google for scraping their content. So it's taking the work that journalists have done and using it to generate those summaries, and media organizations of all sizes will be affected. But I'm particularly concerned, Professor Nicola, about local media.
Do you have any insights or feedback on how this may be affecting them so far? And further to this, do you have any thoughts on how best to regulate the generative engine optimization, because this is rapidly developing alongside search engine optimization, which has traditionally been used to influence how prominently the content appears in search results. So, as you can tell from what I'm saying, Professor, I'm concerned about how these summaries may be influenced or manipulated. This is as much a consumer issue as a media issue and helping journalists to survive, traditional journalists, and I'm concerned that people will be ripped off. So looking forward to your views, Professor Nicola.
Thank you.
Alexandra Geese
Yes, thank you very much for that interesting presentation. I would just like to comment on one thing you said. You said the previous system where Google and the media outlets share the advertising revenue worked for all. I would strongly contest that because the media outlet crisis and the press freedom crisis started when Google started taking half of the revenue that before flew to publishers. I think this is what you showed us.
It shows that media pluralism is slowly being killed. I wouldn't say it's, you know, dying a slow death. It's slowly being killed by Google. And this is going to go further if we don't stop it now. I have talked to publishers, especially local publishers, and they say, well, I know they are breaching copyright.
I know they're taking information locally which only my reporter has reported on, but there's no way I can sue them because they have like thousands of lawyers and I have one person, if I have one person, and I can't have this one person working on that. So the copyright defense does not work here. It only works for really big media houses which usually then cut huge deals with Google. My question is, did you estimate the financial damage? Because you say there's a 40% decrease in referrals to the website basically.
Because this is going to kill our European media diversity completely.
Is there also a 40% decrease in revenue?
Thank you, Professor Lucchi, please for three minutes. We don't have more time.
Nicola LUCCHI
So, thank you for all these questions. I have the impression that we have very few time, so let me put together all these questions because I think they converge on one question: what can we actually do in order to solve this problem? And unfortunately, in the briefing, I was not able to include something that was at that time only a proposal in the UK from the UK Competition and Market Authority.
And so let me close with how to solve in Europe using something that probably can be copied from the United Kingdom proposal. Exactly six weeks ago, the UK Competition and Markets Authority ordered Google to let publishers refuse the use and added a rule that refusing cannot cost them the ranking. So there is no copyright law involved, no need to prove Google broke anything, one designation, one condition. And Europe, I think, has a different instrument capable of producing much the same results without the need of a new law. Google Search is already a designated gatekeeper.
The Digital Market Act already requires fair conditions of access for business users, and publishers are business users. So it also lets the Commission to specify what compliance actually means with binding measures, and this can be done in six months. The Commission is already running one of those on Google Search right now. So I think that probably this is the most effective instrument that we can use in a very short time period because, of course, I mean, technology is going fast. We cannot wait a decision from the competition authority or, I mean, to invent something that is too complex.
We need to act in a very short time period.
Sorry for compressing everything, but I have the impression that there was no space for a full discussion.
Speaker 29
Thank you, Professor Longhin, thank you for our understanding for the situation. I speak in Romanian.
Vasile Dîncu
Ultimul punct pe ordinea de zi este schimbul de opinii cu doamna comisar Marta Kos privind aspectele scutului democrației legate de țările candidate. Sunt încântat să o salut pe doamna comisar pentru extindere. Vreau să îmi cer scuze pentru întârziere, dar aici în Parlamentul European avem opinii și trebuie să ne respectăm reciproc opiniile, inclusiv lungimea lor. Comunicarea comună privind scutul democrației clarifică faptul că acțiunile de combatere a manipulării informațiilor și a ingerințelor străine, pe care comun le numim FIMI, nu se pot opri la granițele UE. Țările candidate, alături de potențialii candidați și partenerii din vecinătatea estică, sunt tot mai des ținta amenințărilor hibride și prezintă, prin urmare, o parte integrantă a strategiei de reziliență a UE.
Așa încât comunicarea prevede mai multe acțiuni, cum ar fi înființarea unui Centru European pentru Reziliență Democratică, sprijinul continuu al Uniunii Europene pentru libertatea presei, jurnalismul independent și rețelele de verificare a faptelor în țările candidate, o implicare mai mare a candidaților în activitatea Observatorului European pentru Media Digitală și a Rețelei Europene pentru Cooperare și Alegeri, utilizarea setului de instrumente FIMI împotriva actorilor care vizează nu numai țările UE, ci și țările candidate.
Vă mulțumim pentru prezență, doamnă comisar. Aveți cuvântul pentru 10 minute.
Marta Kos
Mulțumesc, dear Chair, dear members of the European Parliament. The Democracy Shield has been adopted eight months ago. It is the first ever European comprehensive strategy designed to help protect our democracies. In my work as Enlargement Commissioner, I experience firsthand the importance of our efforts to defend our democratic civic space. Elections are often won on small margins, and in our candidate countries, especially those on the hybrid front line, such as Georgia and Moldova, but also Armenia, this can make a generational difference.
A free and especially fair elections are more than a democratic expression, they matter to the Union's security. And this is why the Democracy Shield places the candidate countries and potential candidates within the remit of actions foreseen for Member States, separately from the rest of international partners.
So the candidates and potential candidates have already a special status in this. This is a strong message about how important our closest neighborhood is for European democratic resilience and security. Geopolitical changes and actions of foreign actors hostile to European integration have exposed several weaknesses of our societies and of our institutions, for example during elections. With examples of Romania or Moldova, we see actually no difference between candidates or Member States. They are attacked alike, simply because they have chosen the EU integration and democracy.
Work of the special committees on the European Democracy Shield is an important contribution to our collective efforts. I'm particularly grateful for your attention and support to the enlargement dimension. The Democracy Shield focuses on protecting the information space, strengthening democratic institutions, reinforcing free and independent media, and promoting societal resilience and participation.
To that end, a range of action is currently being developed to bring together Member States, candidate countries, and potential candidates. First, the European Network of Fact-Checkers is expanding its fact-checking capacity and coverage to enlargement countries. Second, the mandate of the European Digital Media Observatory, EDMO, has been reviewed to develop new independent capabilities to analyze the online information ecosystem, strengthen situational awareness, particularly during elections and crises. Together with our Vice-President Henna Virkkunen, we work on launching a new EDMO hub covering the six Western Balkans in early 2027. Ukraine and Moldova are already covered by the existing EDMO in the EU.
Third point, several enlargement partners are part and playing an increasingly active role in the European Network of Elections, ECNE. This joint work covers several areas of electoral preparedness, ranging from the use of AI in electoral processes to safety in politics, election integrity, and political campaigning. Fourth, the core body of the Democracy Shield is the European Centre for Democratic Resilience, you know it well, which was launched last February.
The first step of the onboarding of enlargement partners in the center took place during the high-level technical meeting last week, 9th of July. If you want more information, I can give it to you what happened there. It will be taken forward at the political level on the 4th of September in the margins of the informal General Affairs Council with ministers from the EU and candidate countries, and this marks really important steps towards direct engagement of candidates with the member states in concrete projects. Fifth, to build a stronger societal resilience, the stakeholder platform will also be part of the Center for Democratic Resilience. It is currently being designed.
Candidate countries and potential candidates will also participate. The platform will facilitate the exchange of knowledge and good practice among all stakeholders, recognizing the crucial role of non-governmental actors in the protection of democracy. Sixth, talking about societal resilience, engagement with young people is our investment in future democratic resilience. Making them aware of threats, involving them in actions protecting democracy and active participatory public life is complementary to other urgent steps we are taking under the Democracy Shield.
To this objective, I held the Youth Policy Dialogue in Moldova at the beginning of June on this topic. So Youth Dialogue is an event where we invite young representatives from the Member States and candidate countries. Seventh, in parallel to the above actions under Democracy Shield, I wanted to walk an extra mile and with my services we are stepping up bilateral engagement with enlargement partners in strengthening their democratic resilience and security. And this work focuses on identifying vulnerabilities and addressing in an integrated way three main areas. First, information integrity.
Second, democratic processes, civil society and media independence. And third, security including hybrid and cyber threats, areas that are closely aligned with the priorities of the European Democracy Shield. Targeted capacity building is being delivered to EU candidate countries and potential candidates to reinforce their resilience to hybrid threats, including information manipulation and hybrid interference.
This is done through regional and bilateral funding geared at long-term support and capacity building, as well as with the TAIEX instrument. TAIEX provides short-term technical assistance and expertise to strengthen public administration and CSDP missions, which support partner countries through security sector reform, rule of law, and civilian or military advisory activities. Dear members of Parliament, the Commission is investing considerable effort into making the European Democracy Shield achieve its goals and into concrete actions with impact, such as the European Center for Democratic Resilience. Over the course of this year, enlargement partners have shown a very strong interest in taking part in our efforts. This is also because democratic resilience is an area where they feel they have a lot to already contribute to us with their knowledge, expertise, ideas, but very often also with first-hand experiences of massive hybrid attacks, like from Moldova, Armenia, or latest from Montenegro.
Candidate countries are in a way springboard towards member states, new tactics are often tested there.
And we see that the more we are successful in the enlargement process, the more the hybrid war or disinformation actions are happening. And one very important part, with important elections coming in many member states next year, we have an interest in accelerating the implementation of the democracy shield both in the EU and also in a candidate country. Our efforts in this direction will continue and we look forward to seeing the first projects where member states and enlargement partners work together and make a real difference on the ground and I count you on your support. The last point I want to stress is actually the whole accession process is built in a way that we are strengthening democracies in candidate countries. We have many activities already ongoing, but now it is important that we really take care about constant exchange between what is happening in the member states and what is happening in the candidate countries.
Some of them are very skilled how to tackle disinformation, the others don't have the structures. So for instance, when Montenegro asked us to help, for instance, very interesting, one day after we have announced that we have built a working group for the preparation of the accession treaty, the next day the number of hybrid attacks went up and we have seen stronger and stronger disinformation against Montenegro's path into the EU. I hope that we will be able to control this much better in the future. We are sending, we call them screening teams, so that see what kind of structures our candidates are having, what is missing. One part of course are the government structures, it's strategic communication, then it is independent media.
Independent media are more and more the question of national security. If you would have asked me 10, 15 years back, and I see some ex-journalists sitting in this room, that we have to support and giving a lot of money to the media, I would say why should we do this? I'm talking about 15 years back. But today, having an independent media, taking care that the integrity of the information will be at a high level, it is important. So I'm happy to be here and to discuss further on.
Vasile Dîncu
Many thanks, Commissioner. Please accept a round of questions of my colleagues. We begin with Ondřej Kolář, EPP. Ondřej.
Ondřej Kolář
Thank you very much. Actually, the Commissioner answered all the questions that I had during her speech, but maybe one small question. You were talking mostly about the candidate countries, but when it comes to the enlargement process, it seems to me that it's harder and harder to explain to people, to citizens of the EU countries, why it is important to enlarge. And I think that a lot of people who already live in the EU are in a way massaged by various fake news when it comes to enlargement, disinformation and all the things and tactics that you have described. So does the Union remember this, or sorry, the Commission remember this, and is there any strategy how to better target our EU citizens and explain to them why enlargement is important to us?
Thank you.
Speaker 33
Thank you very much, Chair. And good that you're here, Commissioner Kos, in this committee.
I'm not so surprised that you already answered a lot of your questions because I know where your heart lies, that is with enlargement and with democracy. And as you rightly said at the end, these two things are so much connected and that is the task that we have to drive that enlargement process. And I strongly believe in that, but we have so many challenges in front of us. And I'm very happy that in our report that we voted here in the committee, enlargement has a very, very important place in the report. So I think let's be honest, we did a good job there.
But now we need to implement that, and in that direction goes my questions. You mentioned Georgia. The situation in Georgia is really, really, really very painful on all the tick-the-box elements of democracy. But there's also another enlarging country that is in a very dire situation, and that is Serbia, when it comes to what you mentioned, the freedom of press, but also civil society.
If you talk to civil society both in Georgia and Serbia it's it's like it's very similar in the sense that they're at a sort of dead end. If if there's not going not anything if nothing is gonna happen they will die they will not be anything. And that would be my first question. When you look to the multi-financial framework and the discussions over there and combine that with the fact that there's no more money coming from the United States to support democracy and all these issues in enlarging countries how can and how should we solve that and where can we help to support that because we need budget to support both media as civil as civil society. And then very concrete to Serbia there will be an election coming up we don't know yet when.
What do we need to do to put the highest guarantee that there is a possibility for free and fair elections? It would be important that you give us some insights there. Thank you.
Csaba Dömötör
Thank you from Patriots for Europe Csaba Dömötör.
Thank you chair let me speak in Hungarian. Tisztelt biztos asszony a többi felszólalótól eltérően én a demokráciavédelem egy másik vonatkozásáról szeretnék beszélni a kisebbségi jogok helyzetéről mégpedig Ukrajna kapcsán. Bővítési biztosként azt gondolom hogy ez is önhöz tartozik felelősségi területként.
Ön számára is ismert, hogy több százezres magyar kisebbség él Ukrajnában, Kárpátalján, és az is ismert lehet, hogy Ukrajnában pár évvel ezelőtt egyes törvények jelentősen szűkítették a nemzeti kisebbségek, így a magyar kisebbségek nyelvhasználati és oktatási jogait. Ez különösen a középfokú oktatást érintette, ahol korlátozták a magyar nyelv használatát az iskolákban. Az új magyar kormány tárgyalásokat folytatott a kijevi vezetéssel, és azt mondják, hogy ennek vannak eredményei. Ha ez így van, akkor az jó hír, de vannak körülmények, amelyek mégis az aggodalom hangját és hangjait hozzák elő. Egyrészt a megállapodás nem terjed ki fontos területekre: kulturális autonómia, parlamenti képviselet, hivatalos névhasználat és a nyelvhasználathoz kapcsolódó szankciók ügye.
Szóval ez a lista szűkebb, mint amit a korábbi magyar kormány képviselt. Másrészt a tárgyalások ellenére az ukrán államnyelvi ombudsman figyelmeztetésben részesített magyar iskolákat, hogy tartsák be a nyelvhasználatot korlátozó intézkedéseket. Tehát a gyakorlatban nem a nyitás jeleit látjuk, hanem inkább a korlátozásokét. Ön egy korábbi nyilatkozatában azt mondta, hogy járt Kárpátalján és a magyar kisebbségek lakta területeken, és rendben vannak a kisebbségi jogok.
Hát amint a példa mutatja, ez nem egészen van így, ez nem teljesen állja meg a helyét. Éppen ezért azt szeretném öntől kérdezni, hogy a csatlakozási tárgyalások során hajlandóak-e foglalkozni a történelmi kisebbségek jogaival, és egyáltalán hogyan fogják ellenőrizni, hogy a vállalásokat az ukrán fél végrehajtja-e? A Bizottság sokat foglalkozik különböző kisebbségi csoportok védelmével, jó lenne, hogyha ezt a figyelmet a történelmi kisebbségek is megkapnák. Köszönöm.
Cristian Terheş
Thank you so much. You were mentioning in your speech that we, you see in the, in certain countries, so to speak, foreign influence. And we've seen that in Moldova, Armenia, and other Western Balkan countries, especially how Russia is using the Orthodox Church there. There are multiple reports even of this Parliament that are outlining that. What, I mean, from your experience now, supervising kind of these elections, what lessons can be learned from, from how Russia is using the religious entity in this case?
Because we need to find a fair balance between religious freedom and belief on one side, and on the other side making sure that we safeguard the democratic process.
Another question is about interconnectivity. Many of these countries that are at different levels or stages in the accession process, they play a key role in interconnecting parts of the world with the European Union. How do you see this component, the role of it in the accession process? Thank you.
Helmut Brandstätter
Thank you, Cristian. From Renew, Helmut Brandstätter.
Mulțumesc and thank you very much, Commissioner Kövesi, for your report. I have four questions. First, the first question, you talked about hybrid war in different countries, in Moldova, Montenegro and others. What does the Commission do to inform people? Do people understand in those countries what hybrid war is, how it is managed, who is doing it, why they are doing it?
So is there kind of information campaign in those countries? I just named those two and if there are others, my question as well. And second, to help media. That's always very difficult. Obviously, media in those countries need help.
On the other hand, it's difficult how you organize that, and of course it has to be transparent. So how is it the plan to help media in these countries? And then, of course, the media situation in Serbia. We know that there are very few independent media. We know about the one TV station N1, and Vučić is doing everything to, yeah, do everything that they don't do what they are doing, a free report.
And so I proposed that we have a kind of a round table in Serbia in September or October with the media, with the government, with the opposition, with journalists, of course. Could you imagine that the Commission would help or would organize something like that? Because I really would say, and it's very simple, with the media landscape as it is now in Serbia, there is no way they can enter the European Union. And my fourth question is about the USA and fake news from the USA. We see what Trump and his family are doing in Albania.
We see how their interests are also especially in Republika Srpska. Obviously they are financing Dodik as well, anyway it looks like that. And I suppose there is also kind of a media campaign from the US in certain countries on the Balkan in order to tell them you don't need Europe, you need US, but it's only about making money for the Trump family. So what can we do to counter those disinformation? My last question, you talked about the resilience center, and we discussed here about the resilience center and also with Commissioner McGrath, and then he said, yeah, it's already started, and obviously it has not started.
Can you give us the latest view where the resilience center is right now and when we really can start cooperating with them, because I think that would be important. Thank you.
Vasile Dîncu
Thank you. We have many colleagues to the list, please answer for the first round. We have many after H-D-I.
Marta Kos
Thank you. Andrej, it's one of the most important points I see and at the same time one of the biggest fears I have.
What I'm talking about, we can do everything in the accession process on the technical level, let's say now with Montenegro coming to the end of the negotiations, we could have a green light from the member states, meaning concluding the accession treaty, but if we will not work with the citizens and with all the EU member states, it could happen that we will not get the final green light. And we have started a year ago, we call it engagement campaign, not a classical communication campaign, engagement campaign in the countries where the support for the EU enlargement is below 50%, especially we are concentrating on France. It is important and I think no enlargement can happen without the member states represented in the Council and it cannot go against the citizens. But sometimes here we have to be careful how the citizens are building their opinion. So how we tackle them that they can get the information, this will be the closer we get to 28 when Montenegro could join if they deliver until the end, the more we will have to invest in this.
I have just met the president of the Social and Economic Council representing trade unions, many other civil society organizations.
We have to work with them. And please, my my call to action to you, it's not enough that we are doing our job here in Brussels. We have all of us, we have to do our work all in the the member states. You are coming from there. And also I would like to see, you know, perhaps sometimes we we go together to those countries which you know, which you know the best.
So thank you for for this question which really speaks about that what is happening through this endangering our democracies is by far not only in our candidates. It is happening also in the in the member states. And this is the main point of this strategy. Thank you. And then Kathleen, indeed, without trust, it is not it is not possible.
We are working on this that we will really increase the amount of money for civil society organizations in the new in the new MFF.
Where we have to be careful and what I am pushing for is that we have to have a system. We have clear priorities in the European Union and when I speak about democracy shield or protecting the democracy, it means that we will have to choose carefully which civil society organizations we are financing or co-financing, meaning that they will do the work which needs to be done, not that they get the money to do something what they think is important. And of course also the, also the evaluation. On, on Serbia, the most important part, we don't know when the elections will be, probably October, November, we'll have parliamentary, we'll have presidential elections. Two important point: about media pluralism and about free and fair elections.
And the last laws which Serbia have to deliver and is delivering is on the REM. It has been built up with a lot of engagement from both sides because exactly of the point you have said. There was no trust anymore on the side of the civil society organization that they told us, "We don't want to come back to work with this government."
So now they have agreed and I hope that soon they will be very very capable of doing their their work. Because of this was not running well in the last year, as you know, I have withheld the payments and these are now the preconditions what Serbia needs to do that we will be able to go to go on. For the question of the Hungarian minority, I'm very happy that the government of Hungary and government of Ukraine have been able to have an agreement on how to protect your minority in Ukraine. And this is now the part of the action plan on minorities. Without this agreement, Hungary would not agree to open cluster one and also this week cluster six.
So there was a huge step forward working on the 11 points from your government and everything what you have mentioned is actually most of them, so the government of Hungary, then you have to talk to your government if you are not happy, but the most of the points the Hungary your government wanted to have is included in this agreement.
And one next point, the accession process and fundamentals negotiations in the fundamental cluster is the best way to get the protection not only your Hungarian minority, but all the minorities. There are more Bulgarians living in Ukraine than Hungarians, meaning in this action plan on minorities, what this is now part of the accession process, Ukraine will have to take care about all the minorities, not just not just one. The question of disinformation and role of the Orthodox Church. We have seen or we see in many countries that a kind of paradox, the Orthodox Church like in Moldova or in Serbia is very trusted institution. But in some cases, the Orthodox Church is also spreading disinformation.
So the first point is that we are also and me personally engaging with representatives of the church. I have met in Moldova representatives of all the churches, the rabbi was there, the bishop was there and the metropolitan from both of Orthodox churches. I have met the representatives of all the churches from Ukraine.
This is ongoing, ongoing work, and I see church as a very important part of the society. You know, we cannot think about civil society engagement without our engagement with them. Thank you for this. Oh, on the interconnectivity, we could talk about this a lot. What is important to see is a big picture.
We have a kind of matrix, if I speak about the interconnection between Europe and Central Asia through countries you know well, Armenia, Azerbaijan, and also Türkiye. The big picture is, first, this would not be possible without peace among, between Armenia and Azerbaijan. Secondly, the first engagement among two countries has been, I'm so happy, the meeting of the civil society organizations, now on both sides, once in Baku, once in Yerevan. So even if the peace is not signed, we can move on, and here we have big plans, and I'm happy about your engagement. You will get more information if you want.
Helmut, you have asked me many questions. So you are one of the journalists probably who know how important the media is and how important it is to inform what is actually the hybrid war.
So we have no special campaign on this, but when we engage with local media, we also ask them to do this work. What is actually hybrid war? You know, it is about cybersecurity. For instance, when we have been helping Armenia, one of the biggest questions has been: will Armenia have electricity on the day of the elections? If there is no electricity, the computers are not working.
So it is not just disinformation. It is not just taking care that the votes are not bought. It is not just going after the oligarchs who have bunches of money in some accounts, and we found many accounts in Moldova. It is about technical equipment and it is also about when the disinformations are there, who is the one who has to go fighting? What I personally have learned a lot, especially on the case of Moldova.
First, we have done before the last parliamentary elections mostly debunking, meaning we have monitoring what is happening and then we answer to this. It is not enough by far.
So this is the modern language, debunking is not enough, we have to do the pre-bunking. So it is not enough that we answer to the disinformation once they occur, we have to follow our own narratives. And there are three most narratives we are following in all the countries. The first narrative is you will lose your sovereignty if you join the EU. Second part is if you accept the European money through the accession process, you will get indebted and you will not be free anymore because you will have to pay the loans back.
And the third very interesting narrative is about the identity. You will lose your identity. And this was a clear case in especially in Moldova, but is also happening now in Montenegro, we see when one third of the population is of Serbian, are Serbs, you know, and we see what kind of disinformations are coming.
So this, you will lose your identity, is a, is actually easy to fight back because we have many small countries like my small country Slovenia, my, my big country Slovenia, which is smaller than Moldova, but the EU is protecting our identity. Since we are a member of the EU, our language is one of 24 official languages in the EU. We are so proud of this. Then, EU is taking care about cultural heritage. And Orthodox churches are part, and every church, all the churches are part of our cultural heritage.
So we are supporting this. And so I went a bit broader. Helmut, helping the, the media, I agree with you and I hope you will be part of our seminar on media which we intend to organize in, in Serbia. The last point you have mentioned, it is MAGA. Sorry, I'm too long, just, just I will finish very shortly.
So MAGA ideologists are claiming what? We have studied this well. They are claiming that European civilization is on a decline. They are claiming that our economy is in decline, that, and that we have no democracy anymore. We have many cases where we have proven that our democracy is working.
One of them is elections in Hungary, where there was the biggest hub for MAGA ideologists. I'm a bit humorous now, but it is very serious. There is a lot of money behind what MAGA ideologists are doing, and we have to be aware that the threats are not coming just from the East, but very often also from the West. So we are aware, aware of this.
Vasile Dîncu
Thank you, and please accept the second round for many questions from our colleagues. For the first, EPP, Tomasz Złotowski.
Tomáš Zdechovský
Yes, thank you, Madam Kos, that you are sharing with us your, your view about current situation, about enlargement. I was totally shocked what I heard from Minister from Serbia, and I, it was not only shock. It was really something what I would like to say, we will take out from Serbia from negotiation if this woman is really Minister. This is something what is absolutely, absolutely non-acceptable.
And this is something what really I felt like someone after the Second World War when he visited Auschwitz, he will say that Hitler was the best solution for the Jewish question. And I think this is something what we have to say and send clear message that no one with this rhetoric will join EU. And this is not only about how to work with citizen. This is about how to work also with politician. And my question is simple.
Do you think what can we do more to give the this country from Balkan an envision, but not bring the country what are corrupted, misusing the fund, supporting Russia and China and not supporting our values, bring to the EU as a Trojan horse. Thank you.
Tonino Picula
Thank you. Tonino Picula.
Thank you, Mr. Chair, Madam Commissioner, thank you for being with us today. Disinformation dominates the political landscape in every single country in the Western Balkans for years, unfortunately. But we must single out one country in particular.
As a standing rapporteur for Serbia, I am particularly concerned about persistent anti-EU rhetoric promoted by pro-government tabloids. There is a striking discrepancy between the Serbian government's stated commitments to EU integration and very toxic narratives disseminated through these media outlets. This has, of course, serious consequences for public opinion. According to recent polls, many citizens express more trust in Russia and China than to the European Union. Moreover, support for EU membership remains comparatively lower than in other candidate countries.
But I would like to ask you about something very related to the issue we discuss. How do you assess the latest pressures on the remaining independent media in Serbia, but also in Montenegro, as you have indicated, under the guise of business operation on the market that are changing the current ownership structure, but most likely also the editorial policy? Thank you.
Fredis Beleris
Thank you. Mr. Beleris.
Ευχαριστώ Πρόεδρε.
Η έκθεση αυτής της επιτροπής καθιστά σαφές ότι οποιαδήποτε παραβίαση των υποψηφίων χωρών ως προς το ευρωπαϊκό κεκτημένο και τις κοινές μας αξίες υπονομεύει τη δημοκρατική ανθεκτικότητα ολόκληρης της Ευρώπης. Η έκθεση της Τουρκίας για το 2023 που ψηφίστηκε στην ολομέλεια του Ιουνίου καταδικάζει το επίσημο casus belli κατά της Ελλάδας, την παράνομη κατοχή ευρωπαϊκού εδάφους στην Κύπρο και την παρεμπόδιση έργων κοινού ευρωπαϊκού ενδιαφέροντος. Ωστόσο στην επίσκεψή σας κύριε Varhelyi στην Άγκυρα 30 Ιουνίου δεν κάνατε καμία αναφορά στις κατάφωρες αυτές παραβιάσεις του διεθνούς δικαίου και τους όρους καλής γειτονίας. Στην Αλβανία κλείσατε μόλις πριν λίγες μέρες τα τρία κεφάλαια. Ο κόσμος στην Αλβανία είναι στους δρόμους, διαμαρτύρονται για τέσσερα βασικά πράγματα: επενδύσεις χωρίς διαφάνεια, χρησιμοποιώντας μαύρο χρήμα, μη σεβόμενοι το ιδιοκτησιακό και το περιβάλλον.
Πώς θα φέρουμε την Τουρκία και την Αλβανία προ των ευθυνών τους αν δεν παρουσιάζουμε τα γεγονότα όπως είναι; Τι μήνυμα στέλνουμε στις άλλες υποψήφιες χώρες; Φοβόσαστε μήπως η στήριξη προς την Αλβανία θεωρηθεί από μέρος της αλβανικής κοινωνίας ως στήριξη προς τον Ράμα και δημιουργήσει εκεί που δεν υπάρχουν αντιευρωπαίους; Ευχαριστώ.
Juan Fernando López Aguilar
Thank you, and finally Mr. López Aguilar, please.
I, of course, I do share many of the points that have already been made, the concerns that have been expressed. But the point I'm making is precisely about the nature of this special committee, Democracy Shield, because when we put in place a special committee, the whole point is delivering conclusions in writing, which are a call for action for the Commission. So when it comes to enlargement, we should bear in mind that certainly many of us are convinced that it's not only about fulfilling the requirements of internal market in the so-called individually merit-based exam for every individual candidate for the accession. It's also about, of course, the rule of law, democracy and fundamental rights framework, which is binding, latest ruling of the European Court of Justice, more than ever before, Article 2 values of the EU, they are binding law for the member states and for the candidates to be. So it's a requirement, but it should also be a requirement, I would like you to assess on the matter, the fulfillment of every point being made by the conclusions which are to be delivered by this committee of us, Democracy Shield, because they should be also binding and the Commission should make sure that they comply with.
Thank you, Chair.
And the Commission should make sure that they comply with, they should also be binding for the candidates to be, for the new members of the European Union. One by one, they should also incorporate the lessons learned of our assessment of the threats against the European idea of democracy, which is enshrined in Article 2, and particularly of the instruments and tools that every individual member state should put in place so that we can all be more resilient and protect the European idea of democracy all together in a more effective way in the foreseeable future. Thank you.
Marta Kos
Thank you for another round of the questions or your opinions. I agree with you, dear Thomas. It was appalling. In the year 2026, I will not repeat what she said. It was appalling that we have heard something like this.
I can't imagine that a minister could further on do her work based on this public, public statement.
I'm always, when we talk to the politicians in the candidates, but also in the member states, you know, every word counts. And, Tom, you have said about the low support for the EU in Serbia, for the EU integration. It is also the results of the statements of the politicians, of what the journalists are writing, what the media are reporting. European Union for many years have been accused of interfering, organizing the protests. This is not what we are doing, and I agree with you completely.
It cannot be that a candidate country is using anti-European rhetoric, especially against EU, especially against you being a rapporteur, especially against those who are critical to what is happening in the country. We have some special requirements from our candidates, also from Serbia. And I still believe, and the European Commission believe, there is a place for Serbia in the European Union. We have seen some small steps to the better when it is about the REM, when they have corrected the judicial amendments, which were really bad.
And I hope, Thomas, you will be also speaking about the CFSP alignment. It was agreed that Serbia will go on with a gradual lifting up the CFSP alignment. So we expect that Serbia will give us a clear signal on which side of the history it will be. We cannot afford to have Trojan horses in the EU. I agree with you completely.
That's why, and this is I am answering also to the last gentleman asking me question, how to take care, Mr. Lopez, that we will not have Trojan horses, Thomas's words.
We are now suggesting special safeguards. At the end of September, I will introduce a European Commission report also on how the new safeguards in the accession treaty of Montenegro will look like, and this accession treaty will be a blueprint for all the later accession treaties. I agree with you, we have to follow and we have to work together with you, with the European Parliament and with all the member states and candidate countries when it is about the resilience of our societies. I think that we will also have to go into the education system. Will we be able to differentiate what is a lie, fake news, and what is the truth?
So we also have to go through a process where enabling not just having independent media, but also us to be more careful what we are believing and what we are not. For me and generally, I will conclude this, many things are expressing how our candidates are approaching the membership status. For me, one of the most important is that actually there is a clear sign: we want to become a member of the EU. Part of this is public rhetoric. The other thing is free and fair elections, it is media pluralism, and it is having democratic institutions.
I was speaking about Moldova before quite a lot. One of the reasons that we have been successful in fighting back disinformation in Moldova were stronger and stronger democratic institutions. Judiciary system represents also those institutions. So it is not just about not just about the the media. Working with our enlargement partners is a two-way process.
Being on the front line of hostile foreign interference and other hybrid threats, enlargement partners bring valuable expertise on first hand that can contribute to the Union's own preparedness and resilience, and their closer involvement therefore creates opportunities for mutual learning. So we have to get away from this scheme that it is us telling them what they have to do and how they have to be. It is also us learning from them what they are they are experiencing. I hope that in the future we will really cooperate well on this very important project, and close to my heart, Kathleen, indeed indeed it is. If we will not protect our democracy, who will do this?
If we will not take care about independent media, who will do this? If we will not cooperate all of us, European institutions and the Member States, who will do this work for us? Nobody will. In my opinion, you know, I have the experiences of living outside the European Union and many of you too, and now being a part of the European Union. There is no better project than it is the European integration and the European Union.
But we have to be very careful what is happening not just around us, but also what is happening inside the European Union. This is one of the most important tasks when I speak about our future, how to protect our democracy.
Vasile Dîncu
Thank you, Commissioner, for this very open dialogue and for not ignoring any question. It's not usually in European Parliament. Thank you.
Carlos Zorrinho
Channel 1. Channel 1. Channel 1. Channel 1. Channel 1.
Channel 1. Channel 1. Channel 1. Channel 1. Channel 1.
Channel 1. Channel 1. Channel 1. Channel 1. Channel 1.
Channel 1. Channel 1. Channel 1. Channel 1. Channel 1.
Channel 1. Channel 1. Channel 1. Channel 1. Channel 1.
Channel 1. Channel 1. Channel 1. Channel 1. Channel 1.
Channel 1. Channel 1. Channel 1. Channel 1.
Ask anything about this video — I analyse its transcript to answer.