Artificial intelligence (AI) has evolved at breakneck speed, becoming a key tool in various fields, from science and technology to art and communication.
However, this question arises: Is AI more useful for the humanities than for the exact sciences? Although it was conceived in the field of mathematics and computing, it has currently stood out more in text writing.
This contrasts with its performance in precise calculations and scientific problem solving, where, paradoxically, it shows limitations. Does AI reflect human thinking more than exact calculation?
In this article we will explore how and why AI seems to be a better fit for the humanities than the sciences.
How is AI evolving in the field of sciences and humanities?
In the 1950s, pioneers such as Alan Turing and John McCarthy laid out the fundamental principles of AI, focused on solving numerical problems and creating systems that “think” like a human.
However, for a long time, these machines were limited to making specific calculations, without understanding the context or interpreting the language. Starting in the 2000s, AI was focused on natural language processing (NLP).
This meant a change in focus: it was no longer just about performing calculations, but about interpreting and generating language, something that requires “humanistic” skills such as understanding context, intention and meaning.
Models such as ChatGPT or Google Gemini are an example of this evolution, bringing AI closer to areas of linguistics and communication, expanding its application and showing an ability to “empathize” and respond in natural language.
AI and the Humanities: Natural Language Processing (NLP)
At its core, NLP is about teaching machines to understand and generate human language, a skill that involves not only grammatical structure, but also the interpretation of context, nuances and emotions.
Tools like ChatGPT and Google Gemini represent notable advances in this area, allowing AI to answer questions, compose texts and even dialogue with users in a coherent and natural way.
The success of AI in NLP lies in its ability to process and learn from large volumes of text, extract language patterns and apply them in conversations or writing.
This allows it not only to deliver information, but also to adapt to the tone and style of the user, something fundamental in humanistic disciplines such as linguistics, philosophy and literature.
This humanistic approach reflects an important evolution: AI no longer just analyzes data in a cold and structured way, but is capable of “interpreting” information, something previously believed to be exclusive to humans.
AI and the Exact Sciences: Mathematics, Physics and Chemistry
Unlike language processing, mathematical calculations and scientific simulations require an accuracy that current AIs, such as ChatGPT or Google Gemini, do not always successfully achieve.
This is, in part, because these models are trained primarily on large volumes of general text and not on specialized or symbolic data essential to the hard sciences.
In mathematics, for example, AI can solve basic problems and follow logical patterns, but it often fails in advanced calculations that require formal rigor and precision.
Similarly, in physics and chemistry, modeling complex systems or solving differential equations involves a level of specialization that language-based AIs are not designed for.
Additionally, lacking a deep understanding of scientific concepts and fundamental rules, they may make logical errors in deductions and simulations.
Comparison of Performance in Humanities and Exact Sciences
Understanding Context (Humanities)
Current AIs, like ChatGPT, are excellent at natural language processing tasks. They are capable of understanding and generating text in complex contexts, capturing emotional, cultural and stylistic nuances.
These capabilities allow applications in humanistic areas such as literature, communication and history, where language, context and interpretation are fundamental.
Precision in Calculations (Exact Sciences)
In exact sciences, where mathematical precision and rigor are crucial, AI has limitations. Although it can solve basic mathematical problems, it tends to make errors in advanced calculations and scientific simulations.
This is because language AIs are trained for textual patterns, not symbolic ones, which affects their performance in disciplines such as physics and chemistry.
Adaptability (Humanities)
In the humanities, AI stands out for its ability to adapt to different writing styles, tones, and perspectives. You can personalize responses and adjust language, simulating contextual empathy that enriches human interaction.
Formal Rigor (Exact Sciences)
The sciences require a rigor that AI does not always meet, since it does not natively process mathematical axioms or scientific laws. To advance in these areas, AI must be integrated with approaches that strengthen its scientific accuracy and reliability.
Can AI be both?
Artificial intelligence, although it shows notable performance in the humanities, does not always fail in calculations or results in exact sciences; There are times when it can provide accurate and useful solutions.
However, in the humanities, there are also times when AI can lack assertiveness or deep understanding.** These “mistakes” should not only be seen as failures, but as opportunities for improvement.
By addressing these limitations, we can move towards artificial intelligence that not only minimizes the margin of error, but also enhances its ability to solve complex problems in both fields.
Thus, AI has the potential to become a truly versatile tool, contributing significantly to both the exact sciences and the humanities in the future.
This post is also available in: