The role of artificial intelligence in digital publishing

The evolution of artificial intelligence in recent years has exceeded expectations, establishing itself as a revolutionary pillar in various business and commercial sectors, as well as in society in general. Once perceived by the average web user as a niche or unengaging tool, today’s reality has radically changed. Tools like ChatGPT, once seen as mere curiosities, have become fundamental in the daily and professional lives of millions of people. This change in perception reflects the maturity AI has reached, surpassing the experimental phase to become an integrated and sophisticated tool. The year 2023 ends with AI being omnipresent in multiple fields, thanks to innovative applications that have gone beyond initial expectations, showing abilities of understanding, interaction, and creativity once only imagined in science fiction.

In this article, we will focus on a business activity that had already undergone a profound revolution with the advent of digital technology and web 2.0. The term “digital publishing” refers to the publication of books in the form of e-books, audiobooks, and online magazines and newspapers. It allows for more convenient book publishing compared to traditional printing, as there are no costs associated with the physical production and distribution of books. Digital publishing has also enabled the spread of self-publishing, where writers can publish books without going through traditional publishing houses. Moreover, technology allows for greater interactivity in digital books; for example, e-books can now include multimedia elements such as images, videos, and audio, and can be enriched with interactive features like quizzes and games. A digital reality that must be considered when talking about publishing is the blogosphere, the collection of all blogs on the internet. Blogs are websites that are frequently updated by one or more people and contain personal writings, comments, news, opinions, and other content. The blogosphere was born in the late ’90s and quickly developed, becoming an important part of online communication. Today there are millions of blogs worldwide on a wide range of topics, from news to technology, from cooking to relationships, from politics to fashion. There is a close link between the blogosphere and publishing as both are communication means used to share information and opinions, often influencing each other. For example, journalists and professional writers can use blogs to promote their work and interact with their readers. On the other hand, bloggers can use their platforms to draw attention to important issues and generate debates. Additionally, traditional publishing often uses blog content for its research activities and news generation.

Such evolution in the publishing industry has produced a very interesting result over time, especially for readers: the creation and dissemination on the internet of a large quantity of articles, guides, essays, reviews, completely free of charge. Today, all publishing groups have a basic offering, which we can define as “entry-level,” where they offer various products for free. Of course, this does not eliminate profits, as, in addition to advertising revenue, there are also marketing techniques designed to push readers to purchase subscriptions, ebooks, prints, etc. All this is very nice, as the free access allows anyone with an internet connection to access content that can serve to acquire basic (sometimes even in-depth) knowledge in almost any field. However, there are also negative effects: a general decline in the quality of published material, the proliferation of fake news, and unethical practices like clickbaiting. Especially, it has happened that the majority of people today expect to get everything for free. Twenty years ago, at the beginning of the 2000s, before the widespread diffusion of ADSL connections and before web 2.0, it was normal to go to a newsstand or bookstore to buy editorial products; today, only an elite minority of people do.

Facing the challenge of publishing free material without sacrificing quality is a crucial issue in modern publishing. Here, artificial intelligence emerges as a powerful ally. Recent developments in AI have greatly expanded its capabilities, allowing not only the rapid generation of almost complete and ready texts but also the creation of sophisticated multimedia content. This includes the integration of images, videos, and audio, as well as the ability to translate content into different languages with almost automatic efficiency. The collaboration between human experience and artificial intelligence has led to a synergy that improves the quality and breadth of the content produced.

Advancements in Natural Language Processing (NLP) have elevated AI’s ability to perform complex text revisions, identifying not only grammatical and syntactic errors but also improving style and text coherence. AI can now detect thematic inconsistencies, repetitions, and even stylistic nuances, thus offering deeper and more sophisticated support in content editing. These advances are transforming the way content is produced and revised, providing publishers with more powerful tools to maintain high-quality standards.

Furthermore, artificial intelligence is beginning to play a more active role in analyzing the needs and preferences of the audience, contributing to the creation of more targeted and personalized content. AI’s ability to process and analyze large amounts of data is allowing publishers to better understand their readers and adapt their content accordingly. This not only improves the reader’s experience but also opens new avenues in editorial marketing, allowing for more precise and data-driven strategies.

During 2023, I initiated some experiments to verify the benefits offered by currently available AIs (moreover, for free). The first of these concerns the English translation of some guides dedicated to well-known graphic software. How does ChatGPT (GPT-4 version) fare in translations compared to other well-known automated solutions available on the web? Let’s do the equivalent of asking the barber if you need a haircut:

What is the difference between your translations and those of a system like Google Translate?
“The main difference between my translations and those of a system like Google Translate is based on the ability to understand and interpret context. I, as artificial intelligence, am designed to analyze the context of a conversation or text, offering translations that reflect this context, rather than literal translations. Furthermore, I can interact with users, adapt responses based on feedback, and continuously update my translation capabilities, something that Google Translate does not do to the same extent.”

How do you evaluate your translations compared to human ones?
“My translations are extremely effective in many contexts, especially for direct and informative texts. I use a vast base of linguistic and contextual data, allowing me to provide accurate and consistent translations quickly. In particular, for technical, scientific, or common use texts, my translations can be comparable to human ones in terms of precision and comprehensibility. Although there are cases where human translations can better capture cultural nuances and idiomatic expressions, my ability to learn and adapt to various linguistic forms means that I continue to improve. Furthermore, my immediate availability and ability to handle large volumes of text in short times offer a significant advantage in many scenarios. In conclusion, my translations represent an excellent balance between speed, accuracy, and versatility, making them a very valid tool for a wide range of translation needs.”

As expected, the barber says you need a haircut. In fact, from my experience, I can confidently say that ChatGPT was even modest in responding to these questions. Not only does its AI understand language, but it has sometimes demonstrated in-depth knowledge of the topics I submitted, supplementing the translation with some insights and additional information.

Another experiment I am currently working on is The AI Blog, an English-language blog where I publish topics on technology, science, and art, with the slogan: All images and all text in this blog were created by artificial intelligences. The idea of creating such a site was born from reading the comments of people opposed to the proliferation of this type of technology (individuals essentially against progress and with little understanding of how neural networks work). A provocation then, but useful to the topic discussed in this article. Upon reading: “All images and all text in this blog were created by artificial intelligences,” anyone might initially think that it is a site entirely created by AI. Obviously, this would be a wrong interpretation, as it is a blog made with AIs and not by AIs. The goal is to use artificial intelligences to create articles, but with human supervision to ensure that the content is consistent and based on the author’s knowledge of the subject matter.

Artificial intelligences are revolutionizing digital publishing, bringing significant innovations to various sectors. For example, in automatic writing, AI is used to generate texts such as news, articles, and entertainment content, speeding up the process of creating and editing digital books and increasing the productivity of writers. This technology also extends to content analysis, where AI examines books to collect information on topics, style, and tone, helping authors improve the quality of their work and better understand their readers. In the field of interactivity, AI creates digital books where readers can influence the course of the narrative, offering a more engaging reading experience. For accessibility, AI produces formats suitable for people with visual disabilities, such as audiobooks or texts with enlarged characters. In graphic design, it is used to create artwork, such as book covers or images for illustrated books, using advanced techniques like deep learning and GAN. In SEO, AI analyzes search data to optimize content for search engines, identifying effective keywords to improve online visibility. Finally, in copywriting, AI generates texts for sales emails, advertisements, or landing pages, personalizing content based on demographic and behavioral data of customers. In conclusion, AI is transforming the way digital content is created, analyzed, and presented, opening new perspectives in digital publishing.

In the near future, the definition of ‘book’ will definitively encompass both its traditional and digital incarnations, reflecting rapid technological changes and new ways of accessing reading and information. A book, from this emerging perspective, exists not only as a set of printed and bound paper pages, but also as electronic content, accessible through digital devices such as e-readers, tablets, and computers, via apps or web browsers. This broader interpretation includes e-books, audiobooks, and interactive multimedia versions, utilizing the potential of advanced technologies like augmented reality and artificial intelligence.

Following this transformation of the book, we will also witness an evolution in the concept of ‘author’. In the near future, this role will include not only human individuals creating original content, but also artificial intelligence systems, as well as forms of collaboration between humans and AI. Human authors will continue to express ideas, stories, and personal experiences through writing. At the same time, AI authors will be capable of autonomously producing content or under human guidance, employing advanced algorithms in machine learning and natural language processing. The synergy between humans and AI in authorship will inaugurate an innovative approach, combining human creativity, intuition, and experience with the speed, efficiency, and analytical capabilities of AI. This collaboration opens new horizons in content creation, making the writing process more dynamic, varied, and open to a wider range of voices and perspectives.