“ChatGPT isn’t sustainable at all. We’re using a number of computational resources involving petabytes of memory and tens of thousands of computational nodes, maybe simply just to ask it to tell us a joke.”
“ChatGPT isn't sustainable at all. We’re using a number of computational resources involving petabytes of memory and tens of thousands of computational nodes, maybe simply just to ask it to tell us a joke.” This is one of the reflections made by Dr Cristina Montañola in the latest episode of the Indefugibles podcast, led by the IQS Department of Ethics and Christian Thought.
With “ChatGPT: artificial intelligence possibilities and limits,” Dr Xavier Casanovas and Dr Oriol Quintana discussed this new artificial intelligence tool that simulates near-human language. Through a debate with computer engineer Dr Cristina Montañola, a professor with the IQS Department of Quantitative Methods, the limits and possibilities of ChatGPT were questioned and some of the debates it is generating were discussed.
Through an informative explanation, Dr Montañola highlighted how ChatGPT works from the point of view of programming as well as for individuals who use it: “it's actually a chatbot that enables making AI accessible to anyone, and has become especially famous because it has achieved high quality and a record number of users: 100 million users in just two months."
The discussion resulted in divergent opinions about whether or not this technology improves the cognitive abilities and skills of people who use it. Dr Quintana questioned its usefulness and purpose: “I doubt that it can improve texts. And what about writing abilities?" Dr Montañola emphasized that “it depends on how you use the tool.” Dr Casanovas, quoting the philosopher Josep Cobo, made a simile between the machine and human strength, and artificial and human intelligence, to pose the question “up to what point will our communicative intelligence decrease?”
The use of the ChatGPT by students, the interests of large technology companies, the “dehumanized” human work behind it, its origin, and the computational challenge concerning the scalability of technology were other leading topics during the conversation.
The hosts of the podcast also highlighted the discussion they organized at IQS for professors, which saw great attendance due to “the expectations and desire to talk about the subject.” Dr Quintana added that scientific workers would have to “have a critical mindset in the face of new technologies,” while Dr Casanovas put an emphasis on “discourses regarding uncritical acceptance of technology, as if the technology has a life of its own that has escaped our control and we are the ones who must adapt.”