Artificial Intelligence (AI) is not new. In one way or another, technologies that apply it have been in use for many years. However, the launch of Dalle2 in March 2022, which raised the possibility that machines can be creative, and ChatGPT in November of the same year, which introduced a structural leap in the application of NLP (Natural Language Processing), have been the triggers for AI to become the trendy phenomenon of 2023.

In reality, AI has a multifaceted nature: it is many things in one. So much so, that the most sophisticated companies refer to it with different terms depending on who their interlocutor is: thus, the concept "Artificial Intelligence" is used with investors, as it is the 'hype' of the season; to attract engineers to develop it, they refer to "Machine Learning"; and they prefer to talk about "Statistics" when they focus on pure code debugging.

Three key elements have led to this situation: the massive amount of data being generated, the processing capacity of this data, and the accessibility for the use and training of open AI models.

This is allowing cases of increasingly disruptive and perfected use of AI to appear at breakneck speed. From virtual TV presenters to political leaders who 'learn' English from one day to the next, through the 'resurrection' of The Beatles or the jury in the US that awarded the realism of a photograph made with Artificial Intelligence.

The communications sector is no stranger to this paradigm shift. In fact, AI is already being used to generate massive audiences in order to monetize them through digital advertising. Never before have consumers had so many possibilities to access content on the Internet, both in social networks and in other repositories and digital media. The quality of the content is another matter.

The major platforms play a decisive role here: Internet search engines and social networks; the latter were born as spaces for users to communicate with each other, but have become true mass media. Due to their own (and legitimate) business model, these platforms are focused on obtaining the maximum benefit from the content that circulates on them, without assuming any editorial responsibility and, in some cases, even endangering the mental health of users, particularly the youngest ones.

In order to maximize audiences and increase profits, the algorithms of some platforms favor free and low-quality content, aimed at capturing clicks, to the detriment of rigorous and in-depth information. They thus secure a competitive advantage as advertising showcases compared to traditional media, but at the price of opening wide the doors to fake news, hate speech and the polarization of democratic societies.

As a result, quality information generated by solvent media loses its momentum in the digital ecosystem. First, because trying to find out the truth is much more costly than fabricating hoaxes. And second, because explaining complex realities with rigor is much less attractive to large audiences than biased, divisive and non-contrasted content.

Prisa Media faces the challenge of AI not as a threat, but as an opportunity. Since our strategy is not based on cost leadership, but on product differentiation to increase our revenues, we will put technology at the service of the best journalism, relieving reporters of lower value-added tasks so that they can focus on quality. 

We build an emotional bond between our newspapers and our readers, based on exclusive news, differential approaches and value-added opinions. That is our brand equity: that readers have to come to us for information and that, satisfied, they are encouraged to subscribe. At the same time, we aim to offer our advertisers the widest and best quality audience to meet their commercial, institutional and reputational objectives.

We are not naïve; we know the dangers lurking in the communication sector. AI will undoubtedly have a multiplying effect on the free circulation of 'fake news' and polarizing content, with the connivance of networks and search engines. A real weapon of mass destruction for democracies if not properly managed and controlled.

It is therefore urgent that governments regulate the digital information market in at least two ways.

First, to require the traceability of the content on which Artificial Intelligence feeds, in order to ensure both the reliability of the information on which it is based and its respect for intellectual property. To do otherwise would mean allowing AI to feed itself opaquely and free of charge from the media and then to compete with them.

And second, to align the legal limits of AI with the sustainability criteria that govern governments and companies. Yes, the ESG axes (environment, society and governance), aimed at encouraging companies to seek not only economic profit, but also the common good. But there is also another vector that we must not forget: the sustainability of democracies.

Increasingly, compliance with ESG criteria is conditioning business decisions. And yet, these same companies, when they need to attract customers, have no qualms about investing in media and supports where their commercial messages, which claim to be friendly and positive, circulate alongside misinformation, hate speech and anti-democratic fanaticism.

Taxonomizing content creators who work for coexistence (and those who undermine it) and incorporating the commitment to democracy into sustainability criteria would allow advertisers to better objectify their investment priorities. And for the media committed to the common good, to improve their profitability in order to continue betting on quality journalism. Because the world of communication cannot become a lawless territory in which those who contribute to consolidating democracies and those who fight to destroy them coexist on an equal footing.