AI won’t replace quality journalism, but sector needs safeguarding, says socialist think tank head

Despite the rapid rise in Artificial Intelligence and large social media platforms, the President of the Foundation for European Progressive Studies (FEPS) Maria João Rodrigues believes that quality journalism still has a role to play in Europe’s future.

The former S&D Group Vice President spoke with Euractiv’s Christoph Schwaiger on the side-lines of the Stars4Media day event on AI and EU enlargement, and called for concrete measures to be taken now, rather than later, to safeguard European journalism. 

CH: Journalists are known to be sceptical of professional advice from non-journalists. What would you say is your motivation to improve the future of the media sector?

MJR: I’ve been in European politics for quite long periods with different hats. As minister, then as parliamentarian, then working in the European Commission and the European Council. So I made the tour. Now Europe is preparing for the next phase, so the next five years. And I can tell you that as time goes by, the more I conclude that the role of media is completely crucial.

Society has to decide and discuss what it wants during this phase. If we want a good discussion we need to count on reliable sources of information. For democratic societies the role of the media is absolutely central. But it has to be a media environment with freedom, pluralism, and quality.

Important initiatives were introduced over the last years. Most recently the European Media Freedom Act and the AI Act. A long sequence of instruments has been adopted but it’s not enough. We’re going through a very challenging period – war in Europe, competing global powers, and competing narratives. We’ve also decided, rightly so, to support countries in Eastern Europe. The media also has a role to play in this context.

CH: There are a bunch of platforms serving up content to people. Now AI can provide up-to-date answers to people’s questions. Given this abundance of information and resources, what role should the European media play?

MJR: The digital era radically changed the way people use media. Access to knowledge is no longer dependent on time, place, and space.

In the Web 2.0 era people started creating content themselves. Now platforms have completely changed the way consumers interact with producers. Today we’re faced with algorithms and AI. We need to recognise AI will replace some capabilities of human intelligence. But just some.

Within seconds, ChatGPT can answer most of your questions. But human intelligence can’t be replaced in some other tasks. Such as choosing what we want to know. This is a human decision. And then framing what you want to know.

This is why quality journalism will remain a crucial activity for human societies. I like to count on reliable quality professionals to help me to choose what I should know.

When I wake up I look at the news and I seek reliable brands. By the way, I use Euractiv. Because I’m interested in having a team of people helping me identify what has happened over the last hours not just where I’m living but across Europe and across the world. I consider myself to be an active citizen and also a politician so I need to use a lot of information.

If we really want quality journalism then we need to take steps, otherwise it becomes at risk.

CH: What kind of steps?

MJR: We need to assess the new context journalism is operating in. Democratic debate is under pressure from actors that try to disrupt, instrumentalise, and manipulate it.

We can’t be naive about this. I have no doubt that Putin’s regime is doing this in a systematic way regarding debates in each European country.

Secondly, citizens are consuming much more information, which is good. But powerful platforms, mainly in America, are framing most of the debates in Europe. They’ve also amassed large amounts of data creating personal profiles of their users. This is the new raw material. Big data.

These same platforms also have access to big financial sources via advertising which is part of their business model. They can use all this to decide or frame European debates because they know what people are looking for. Then they can target them with an advertiser’s messages. The risk is that the way we perceive reality becomes framed by this business model.

If we each have our own completely different perception of reality, that would greatly impact democracy. We saw this in an extreme case in the US elections. I hope this won’t happen in Europe, but it’s starting to happen. Let this be my wakeup call.

We’ve started implementing legislation, starting with GDPR and then copyright laws. Then came the Digital Services Act (DSA) and the Digital Markets Act. Now we’re working on building European data spaces which will allow citizens to regain control of their own data.

All this is important but not enough. Over the next five years we need to work with two legs – one is legislation and the other leg is capacity building with an active industrial and innovation policy.

We need to create the conditions that will enable platforms to emerge in the private media sector. They could be on niche topics but there’s space for cross-national platforms with automatic translations in different languages. It will enable cooperation between different media companies and it allows for cost sharing. Running these platforms is costly as is the development of AI for different purposes.

Public media also needs improvement. An idea floating around is the creation of a European platform to share news provided by public media to enable citizens to follow the public debates happening in other countries in a language of their choice.

Creating this means creating a true European public sphere. This would create a completely different way for Europeans to understand each other. They’ll find out they have a lot of problems in common. It will also create a strong European identity.

In terms of industrial policy, there’s a full toolbox. You can start with public procurement, tax incentives, and state aid. But you have to be careful, because you have to remain transparent and protect pluralism. State aid can’t be used to manipulate information. Outlets need to maintain their editorial discretion.

On top of this we need large scale training on how to use new media, for consumers and professionals alike. For instance, teaching journalists how to make the best use of AI rather than being replaced by AI.

CH: Two quick questions to wrap up. Firstly, various participants in the discussions during the AI and enlargement event expressed concerns about social media platforms. The US House just passed a bill that could ban TikTok in America. Is this an approach we should consider in Europe or is it too extreme?

MJR: There is cause for concern about TikTok. Everyone knows TikTok is extremely successful, even in Europe, as it’s been designed to hook your attention and meet many angles of curiosity.

Nevertheless, the key algorithms TikTok uses are produced in China and are not disclosed. Bear in mind that social networks like that collect gigantic amounts of information on users. What guarantee do we have that all this data won’t be used for other purposes?

Behaviours and choices could be influenced in European democratic elections. Ideally we should have the possibility for European citizens to choose among different social networks. I think this is a solution.

CH: And lastly, on your point on state aid. What would happen in places like Hungary? How would we ensure that a citizen’s own money goes into the bank accounts of media outlets that genuinely contribute to the wider European project?

MJR: We need to have full transparency on state aid. We need to ensure state aid is readily available to support pluralism. This is not at all the case in Hungary. It’s not transparent. It’s not plural. It’s used for manipulation. It’s the opposite of what we should have. That’s my reply.

The interview has been edited for clarity and brevity.

[By Christoph Schwaiger I Edited by Brian Maguire | Euractiv’s Advocacy Lab ]

Read more with Euractiv

Council of Europe AI treaty does not fully define private sector's obligations

A draft treaty to protect human rights, democracy, and the rule of law, agreed at the Council of Europe (CoE) on Thursday (14 March), leaves it up to countries to decide how to include the private sector in the development of artificial intelligence (AI).