What is behind semiology that LLM and AI need to embrace? The answer might be a little nerdy… Hang on!

For those who need a refresher… Here are some of the basics:

LLM is: “A large language model (LLM) is a type of artificial intelligence (AI) program that can recognize and generate text, among other tasks. LLMs are trained on huge sets of data — hence the name "large."… In simpler terms, an LLM is a computer program that has been fed enough examples to be able to recognize and interpret human language or other types of complex data.

Semiology is: “Semiology is defined as the science that studies verbal and nonverbal signs, as well as sign-using behaviour, similar to how linguistics studies verbal signs. It focuses on the study of all types of signs, particularly language and its relationship to other sign systems.”

Umberto Eco “argued that cultural artifacts are "open works" (ring a bell?) capable of multiple interpretations, each shaped by the cultural and historical contexts of both the creator and the viewer (Eco, 1989). Eco's theory emphasizes the interpretative flexibility of artifacts, suggesting that each interaction with an artifact can yield new meanings depending on the interpretive strategies of the observer.” Which is also a challenge when referring to open data.

So, what does this have to do with digital? Well, everything because: “…according to the Palo Alto model (see image), the successes in communication between individual will be given because they communicate in the same code, which is not altered within the canal; because the receiver's situation is taken into account; the framework in which the communication happens is analysed; digital communication is in line with analog communication; the punctuation of the sequence of events is well defined and the communicator has its receiver. Otherwise, communication between both individuals fails. Which beg to ask the question: can LLM and AI substitute the same type of variables that would alter interpersonal relationships? And if that is the case for interpersonal interaction, how can LLM and Ai fill-in the gaps in a digital environment?

Let us dig a little further: In 2022, Berlanga-Fernández and Reyes explored the digital approach to semiotics, examining how digital tools and methodologies are applied to the study of signs and meanings within various texts and contexts. Their analysis highlights the evolving landscape of semiotic research in the digital era, emphasizing the integration of technological advancements with traditional theoretical frameworks.

"These studies consider not only the artifact itself but also the digital interface and metadata as part of the semiotic system, which shapes the user's interpretation. Furthermore, theories from cultural studies, such as those proposed by Stuart Hall, argue that cultural identities and meanings are continuously negotiated through cultural practices, including the creation, display, and interpretation of artifacts (Hall, 1997). This perspective is crucial in understanding how cultural heritage artifacts function as symbols within ongoing cultural dialogues and identity formations.”

So, I would say to the AI industry: Don't forget the people behind the data, and moreover, make sure to hire really good semiologist!