How artificial intelligence has already changed the rules of the internet

GenevaSince its popularization in the late 1990s, the internet has established itself as the primary means of digital communication. With the emergence of artificial intelligence in recent years, however, the internet has undergone a paradigm shift, offering users a wide range of new features: from new ways to search for information to personalized shopping. However, these new tools entail a series of risks that must be taken into account in order to browse safely.

"Before we searched, now we ask"

One of the main features of the Internet is the access it provides to information. Search engines are the systems that organize it, so that users can easily find what they are looking for. During the early years of the Internet, these search engines were simple directories that classified websites by topic. This system was very primitive and had limitations that did not always allow access to what was sought. With the arrival of more complex search engines, such as those of Yahoo! and Google, information began to be indexed more efficiently, allowing the user to reach the desired information easily and accurately.

With AI we have entered a new era of information search. on-lineNew "smart" tools allow users to interact with new search engines in a more natural way. "We are witnessing a profound change. We have gone from the search engine to the assistant," explains Antonio Pita, professor of Computer Science, Multimedia and Telecommunications Studies at the Open University of Catalonia (UOC), adding that "before we searched, now we ask."

New AI systems synthesize all the internet information they've been trained with and provide a more or less accurate answer to the question asked. However, using systems like ChatGPT to perform searches can involve some risks. These models are prone to generating false information (known as hallucinations), and often have incorrect or nonexistent sources. Therefore, it's important to always compare information with other sources.

Don't believe everything you see

The advancement of generative AI, which is capable of semi-autonomously creating text or images, has a major impact on the content found on the internet. "We are moving towards an internet where much of the content will be created by machines," warns Pita, who asserts that the quality of that content will become the main differentiating factor.

In this context, the role of human professionals may be challenged. Pita comments that "where there is a different perspective, a well-told story, or a well-founded opinion, something that moves, surprises, or simply makes people think, is where humans will make the difference." Specifically, the advancement of AI has a significant impact on the media. "They will have to reinvent themselves because it's no longer enough to simply inform. They must offer context, interpretation, and trust, and differentiate themselves, because if everything sounds the same, no one will listen," believes this expert.

Among the major challenges facing the media is the fight against disinformation, where AI serves as a driving force behind the promotion and dissemination of this type of content. However, on the other hand, artificial intelligence can also provide a tool to detect manipulation or trace the origin of information and thus be able to verify it. "If this era should teach us anything, it's that we can't believe everything we see or everything we feel," warns Pita, who recommends not leaving everything in the hands of technology.

More sensory and personalized purchases

Artificial intelligence has also brought about a significant change in the way users shop online. In addition to one-click purchases, we must also consider personalized ads based on profiles created by intelligent systems based on our preferences. These agents use the information we leave online to more accurately profile our tastes and promote content that matches our personality. According to Pita, the development of other immersive technologies will allow brands to generate "more sensorial, personalized, and efficient" virtual shopping experiences. We must therefore question whether the internet knows more about us than we do about ourselves.

Learning adversaries

The increasing complexity of the internet has made cybersecurity an even more important aspect in preventing attacks that jeopardize the functioning of some digital systems, as well as the data circulating through them. In this regard, AI is a double-edged sword. On the one hand, it allows for the creation of more sophisticated attacks, and on the other, it also makes it possible to prevent them more efficiently than before. "AI forces us to coexist with a new type of adversary: one that also learns," emphasizes the UOC professor, who suggests promoting a culture of cybersecurity among the population to address these new risks.

A regulation in tow

New technologies based on artificial intelligence pose a series of legal challenges that legislators have been grappling with for years. The main problem is that technology advances rapidly and legislation is slow. Along these lines, the European Union recently implemented the AI law, known as AI Act, which establishes the regulatory foundations for this technology. However, the challenge is ongoing to adapt to the changes occurring in the digital world of the internet. "AI forces us to reflect not only on what we can do, but on what we should do. And that won't be decided by the models, it will be decided by us," Pita concludes.